Image-Captioning.md

Image Captioning

Title	Date	Abstract	Comment	CodeRepository
Mindstorms in Natural Language-Based Societies of Mind	2026-03-11	Show Both Minsky's "society of mind" and Schmidhuber's "learning to think" inspire diverse societies of large multimodal neural networks (NNs) that solve problems by interviewing each other in a "mindstorm." Recent implementations of NN-based societies of minds consist of large language models (LLMs) and other NN-based experts communicating through a natural language interface. In doing so, they overcome the limitations of single LLMs, improving multimodal zero-shot reasoning. In these natural language-based societies of mind (NLSOMs), new agents -- all communicating through the same universal symbolic language -- are easily added in a modular fashion. To demonstrate the power of NLSOMs, we assemble and experiment with several of them (having up to 129 members), leveraging mindstorms in them to solve some practical AI tasks: visual question answering, image captioning, text-to-image synthesis, 3D generation, egocentric retrieval, embodied AI, and general language-based task solving. We view this as a starting point towards much larger NLSOMs with billions of agents-some of which may be humans. And with this emergence of great societies of heterogeneous minds, many new research questions have suddenly become paramount to the future of artificial intelligence. What should be the social structure of an NLSOM? What would be the (dis)advantages of having a monarchical rather than a democratic structure? How can principles of NN economies be used to maximize the total reward of a reinforcement learning NLSOM? In this work, we identify, discuss, and try to answer some of these questions.	publi... published in Computational Visual Media Journal (CVMJ); 9 pages in main text + 7 pages of references + 38 pages of appendices, 14 figures in main text + 13 in appendices, 7 tables in appendices	None
CodePercept: Code-Grounded Visual STEM Perception for MLLMs	2026-03-11	Show When MLLMs fail at Science, Technology, Engineering, and Mathematics (STEM) visual reasoning, a fundamental question arises: is it due to perceptual deficiencies or reasoning limitations? Through systematic scaling analysis that independently scales perception and reasoning components, we uncover a critical insight: scaling perception consistently outperforms scaling reasoning. This reveals perception as the true lever limiting current STEM visual reasoning. Motivated by this insight, our work focuses on systematically enhancing the perception capabilities of MLLMs by establishing code as a powerful perceptual medium--executable code provides precise semantics that naturally align with the structured nature of STEM visuals. Specifically, we construct ICC-1M, a large-scale dataset comprising 1M Image-Caption-Code triplets that materializes this code-as-perception paradigm through two complementary approaches: (1) Code-Grounded Caption Generation treats executable code as ground truth for image captions, eliminating the hallucinations inherent in existing knowledge distillation methods; (2) STEM Image-to-Code Translation prompts models to generate reconstruction code, mitigating the ambiguity of natural language for perception enhancement. To validate this paradigm, we further introduce STEM2Code-Eval, a novel benchmark that directly evaluates visual perception in STEM domains. Unlike existing work relying on problem-solving accuracy as a proxy that only measures problem-relevant understanding, our benchmark requires comprehensive visual comprehension through executable code generation for image reconstruction, providing deterministic and verifiable assessment. Code is available at https://github.com/TongkunGuan/Qwen-CodePercept.	Accepted by CVPR2026	Code Link
AutoViVQA: A Large-Scale Automatically Constructed Dataset for Vietnamese Visual Question Answering	2026-03-11	Show Visual Question Answering (VQA) is a fundamental multimodal task that requires models to jointly understand visual and textual information. Early VQA systems relied heavily on language biases, motivating subsequent work to emphasize visual grounding and balanced datasets. With the success of large-scale pre-trained transformers for both text and vision domains -- such as PhoBERT for Vietnamese language understanding and Vision Transformers (ViT) for image representation learning -- multimodal fusion has achieved remarkable progress. For Vietnamese VQA, several datasets have been introduced to promote research in low-resource multimodal learning, including ViVQA, OpenViVQA, and the recently proposed ViTextVQA. These resources enable benchmarking of models that integrate linguistic and visual features in the Vietnamese context. Evaluation of VQA systems often employs automatic metrics originally designed for image captioning or machine translation, such as BLEU, METEOR, CIDEr, Recall, Precision, and F1-score. However, recent research suggests that large language models can further improve the alignment between automatic evaluation and human judgment in VQA tasks. In this work, we explore Vietnamese Visual Question Answering using transformer-based architectures, leveraging both textual and visual pre-training while systematically comparing automatic evaluation metrics under multilingual settings.		None
MUNIChus: Multilingual News Image Captioning Benchmark	2026-03-11	Show The goal of news image captioning is to generate captions by integrating news article content with corresponding images, highlighting the relationship between textual context and visual elements. The majority of research on news image captioning focuses on English, primarily because datasets in other languages are scarce. To address this limitation, we create the first multilingual news image captioning benchmark, MUNIChus, comprising 9 languages, including several low-resource languages such as Sinhala and Urdu. We evaluate various state-of-the-art neural news image captioning models on MUNIChus and find that news image captioning remains challenging. We also make MUNIChus publicly available with over 20 models already benchmarked. MUNIChus opens new avenues for further advancements in developing and evaluating multilingual news image captioning models.	Accep... Accepted to LREC 2026 (The Fifteenth biennial Language Resources and Evaluation Conference)	None
Fighting Hallucinations with Counterfactuals: Diffusion-Guided Perturbations for LVLM Hallucination Suppression	2026-03-11	Show While large vision-language models (LVLMs) achieve strong performance on multimodal tasks, they frequently generate hallucinations -- unfaithful outputs misaligned with the visual input. To address this issue, we introduce CIPHER (Counterfactual Image Perturbations for Hallucination Extraction and Removal), a training-free method that suppresses vision-induced hallucinations via lightweight feature-level correction. Unlike prior training-free approaches that primarily focus on text-induced hallucinations, CIPHER explicitly targets hallucinations arising from the visual modality. CIPHER operates in two phases. In the offline phase, we construct OHC-25K (Object-Hallucinated Counterfactuals, 25,000 samples), a counterfactual dataset consisting of diffusion-edited images that intentionally contradict the original ground-truth captions. We pair these edited images with the unchanged ground-truth captions and process them through an LVLM to extract hallucination-related representations. Contrasting these representations with those from authentic (image, caption) pairs reveals structured, systematic shifts spanning a low-rank subspace characterizing vision-induced hallucination. In the inference phase, CIPHER suppresses hallucinations by projecting intermediate hidden states away from this subspace. Experiments across multiple benchmarks show that CIPHER significantly reduces hallucination rates while preserving task performance, demonstrating the effectiveness of counterfactual visual perturbations for improving LVLM faithfulness. Code and additional materials are available at https://hamidreza-dastmalchi.github.io/cipher-cvpr2026/.	CVPR 2026	Code Link
Grounding Synthetic Data Generation With Vision and Language Models	2026-03-10	Show Deep learning models benefit from increasing data diversity and volume, motivating synthetic data augmentation to improve existing datasets. However, existing evaluation metrics for synthetic data typically calculate latent feature similarity, which is difficult to interpret and does not always correlate with the contribution to downstream tasks. We propose a vision-language grounded framework for interpretable synthetic data augmentation and evaluation in remote sensing. Our approach combines generative models, semantic segmentation and image captioning with vision and language models. Based on this framework, we introduce ARAS400k: A large-scale Remote sensing dataset Augmented with Synthetic data for segmentation and captioning, containing 100k real images and 300k synthetic images, each paired with segmentation maps and descriptions. ARAS400k enables the automated evaluation of synthetic data by analyzing semantic composition, minimizing caption redundancy, and verifying cross-modal consistency between visual structures and language descriptions. Experimental results indicate that while models trained exclusively on synthetic data reach competitive performance levels, those trained with augmented data (a combination of real and synthetic images) consistently outperform real-data baselines. Consequently, this work establishes a scalable benchmark for remote sensing tasks, specifically in semantic segmentation and image captioning. The dataset is available at zenodo.org/records/18890661 and the code base at github.com/caglarmert/ARAS400k.		Code Link
Image Captioning via Compact Bidirectional Architecture	2026-03-10	Show Most current image captioning models typically generate captions from left-to-right. This unidirectional property makes them can only leverage past context but not future context. Though refinement-based models can exploit both past and future context by generating a new caption in the second stage based on pre-retrieved or pre-generated captions in the first stage, the decoder of these models generally consists of two networks~(i.e. a retriever or captioner in the first stage and a captioner in the second stage), which can only be executed sequentially. In this paper, we introduce a Compact Bidirectional Transformer model for image captioning that can leverage bidirectional context implicitly and explicitly while the decoder can be executed parallelly. Specifically, it is implemented by tightly coupling left-to-right(L2R) and right-to-left(R2L) flows into a single compact model to serve as a regularization for implicitly exploiting bidirectional context and optionally allowing explicit interaction of the bidirectional flows, while the final caption is chosen from either L2R or R2L flow in a sentence-level ensemble manner. We conduct extensive ablation studies on MSCOCO benchmark and find that the compact bidirectional architecture and the sentence-level ensemble play more important roles than the explicit interaction mechanism. By combining with word-level ensemble seamlessly, the effect of sentence-level ensemble is further enlarged. We further extend the conventional one-flow self-critical training to the two-flows version under this architecture and achieve new state-of-the-art results in comparison with non-vision-language-pretraining models. Finally, we verify the generality of this compact bidirectional architecture by extending it to LSTM backbone. Source code is available at https://github.com/YuanEZhou/cbtic.		Code Link
RubiCap: Rubric-Guided Reinforcement Learning for Dense Image Captioning	2026-03-10	Show Dense image captioning is critical for cross-modal alignment in vision-language pretraining and text-to-image generation, but scaling expert-quality annotations is prohibitively expensive. While synthetic captioning via strong vision-language models (VLMs) is a practical alternative, supervised distillation often yields limited output diversity and weak generalization. Reinforcement learning (RL) could overcome these limitations, but its successes have so far been concentrated in verifiable domains that rely on deterministic checkers -- a luxury not available in open-ended captioning. We address this bottleneck with RubiCap, a novel RL framework that derives fine-grained, sample-specific reward signals from LLM-written rubrics. RubiCap first assembles a diverse committee of candidate captions, then employs an LLM rubric writer to extract consensus strengths and diagnose deficiencies in the current policy. These insights are converted into explicit evaluation criteria, enabling an LLM judge to decompose holistic quality assessment and replace coarse scalar rewards with structured, multi-faceted evaluations. Across extensive benchmarks, RubiCap achieves the highest win rates on CapArena, outperforming supervised distillation, prior RL methods, human-expert annotations, and GPT-4V-augmented outputs. On CaptionQA, it demonstrates superior word efficiency: our 7B model matches Qwen2.5-VL-32B-Instruct, and our 3B model surpasses its 7B counterpart. Remarkably, using the compact RubiCap-3B as a captioner produces stronger pretrained VLMs than those trained on captions from proprietary models.		None
Improving Large Vision-Language Models' Understanding for Flow Field Data	2026-03-10	Show Large Vision-Language Models (LVLMs) have shown impressive capabilities across a range of tasks that integrate visual and textual understanding, such as image captioning and visual question answering. These models are trained on large-scale image and video datasets paired with text, enabling them to bridge visual perception and natural language processing. However, their application to scientific domains, especially in interpreting complex field data commonly used in the natural sciences, remains underexplored. In this work, we introduce FieldLVLM, a novel framework designed to improve large vision-language models' understanding of field data. FieldLVLM consists of two main components: a field-aware language generation strategy and a data-compressed multimodal model tuning. The field-aware language generation strategy leverages a special-purpose machine learning pipeline to extract key physical features from field data, such as flow classification, Reynolds number, and vortex patterns. This information is then converted into structured textual descriptions that serve as a dataset. The data-compressed multimodal model tuning focuses on LVLMs with these generated datasets, using a data compression strategy to reduce the complexity of field inputs and retain only the most informative values. This ensures compatibility with the models language decoder and guides its learning more effectively. Experimental results on newly proposed benchmark datasets demonstrate that FieldLVLM significantly outperforms existing methods in tasks involving scientific field data. Our findings suggest that this approach opens up new possibilities for applying large vision-language models to scientific research, helping bridge the gap between large models and domain-specific discovery.	Accep... Accepted by Machine Intelligence Research	None
VIVECaption: A Split Approach to Caption Quality Improvement	2026-03-08	Show Caption quality has emerged as a critical bottleneck in training high-quality text-to-image (T2I) and text-to-video (T2V) generative models. While visual language models (VLMs) are commonly deployed to generate captions from visual data, they suffer from hallucinations, poor compositional reasoning, and limited fine-grained understanding, resulting in misaligned image-caption pairs that degrade downstream model performance. This technical report introduces VIVECaption, a systematic two-sided approach to caption quality improvement. We first establish a comprehensive taxonomy of caption evaluation metrics, distinguishing between "universal" and "instance-grounded" metrics, with the ultimate goal of showcasing the use-cases and tradeoffs between different caption quality metrics. We then use this language to describe our two-sided approach to caption quality improvement: (1) a gold-standard dataset creation methodology using stratified sampling and (2) a model alignment strategy encompassing context alignment and parameter-level finetuning using SFT. We demonstrate our methodology on open-source models, focusing on structured caption formats that enable better parsing and downstream utilization. We ultimately show that using a finetuned character detection model in an image captioning pipeline significantly improves holistic image-caption alignment quality. Our work addresses the growing need for high-quality "vegan" training data in enterprise AI development, providing practical solutions for teams seeking to improve caption-image alignment without relying on potentially copyright-protected web-scraped content.		None
Multi-modal, Multi-task, Multi-criteria Automatic Evaluation with Vision Language Models	2026-03-07	Show Vision-language models (VLMs) have shown impressive abilities across a range of multi-modal tasks. However, existing metrics for evaluating the quality of text generated by VLMs typically focus on an overall evaluation for a specific task, such as image captioning. While the overall evaluation is essential for any task, the criteria prioritized can differ depending on the task, making it challenging for current metrics to adapt to multi-task scenarios. To address this limitation, we propose HarmonicEval, a reference-free comprehensive evaluation metric that aggregates criterion-wise scores to produce the overall score in a bottom-up manner. Furthermore, to assess the generalizability of automatic evaluation metrics in multi-task scenarios, we construct the Multi-task Multi-criteria Human Evaluation (MMHE) benchmark, which comprises 18,000 expert human judgments across four multi-modal tasks. Our experiments demonstrate that HarmonicEval achieves higher correlations with human judgments than conventional metrics while providing numerical scores for each criterion. Project page: https://stjohn2007.github.io/MMHE_project/		Code Link
DeepSight: Bridging Depth Maps and Language with a Depth-Driven Multimodal Model	2026-03-06	Show Multimodal large language models (MLLMs) have achieved impressive performance across various tasks such as image captioning and visual question answer(VQA); however, they often struggle to accurately interpret depth information inherent in visual data. In this work, we introduce DeepSight, the first dedicated depth MLLM designed to enhance three-dimensional scene understanding. Unlike conventional methods that align RGB image encodings with text, our approach takes advantage of the unique characteristics of depth images: single-channel grayscale images where the pixel values directly reflect depth cues to improve spatial reasoning. To address challenges associated with limited depth data and the inadequacy of simple channel replication, we construct a novel depth image-text pair dataset and a depth instruction dataset. Depth maps are generated from visual images using the GLPN model, and GPT-4 is employed to curate corresponding depth instructions, an approach validated by LLaVA. Additionally, we modify the ViT encoder in CLIP to incorporate local object information, thereby capturing the subtle continuous variations of depth more effectively. To evaluate the performance of our model, we develop a comprehensive depth question answer benchmark based on existing depth image datasets, which rigorously assesses understanding in typical depth map scenarios. Experimental results demonstrate that DeepSight significantly enhances depth perception and downstream task performance, marking a substantial step forward in multimodal three-dimensional understanding.		None
Multimodal Behavior Tree Generation: A Small Vision-Language Model for Robot Task Planning	2026-03-06	Show Large and small language models have been widely used for robotic task planning. At the same time, vision-language models (VLMs) have successfully tackled problems such as image captioning, scene understanding, and visual question answering. In this work, we combine these two approaches by deploying a compact, open-source multimodal model to generate behavior trees for robotic task planning. The main obstacle to achieving this goal is the lack of an existing dataset that links visual observations and instructions to executable behavior trees. We propose a method to construct such a dataset starting from existing robotic episodes (i.e., Open X-Embodiment), in which a large model serves as a teacher in a multi-stage generation pipeline. We use this dataset to fine-tune VLMs ranging from 500M to 4B parameters via parameter-efficient fine-tuning (PEFT). The generated behavior trees, compatible with the BehaviorTree.CPP library, are evaluated both offline, using structural and lexical metrics, and online through the execution of household tasks in a state-of-the-art embodied simulator. Our results demonstrate that our fine-tuned 4B-parameter VLM approaches the performance of state-of-the-art closed-source models, achieving an 87% success rate while requiring only a fraction of the computational resources.		None
VisionPangu: A Compact and Fine-Grained Multimodal Assistant with 1.7B Parameters	2026-03-05	Show Large Multimodal Models (LMMs) have achieved strong performance in vision-language understanding, yet many existing approaches rely on large-scale architectures and coarse supervision, which limits their ability to generate detailed image captions. In this work, we present VisionPangu, a compact 1.7B-parameter multimodal model designed to improve detailed image captioning through efficient multimodal alignment and high-quality supervision. Our model combines an InternVL-derived vision encoder with the OpenPangu-Embedded language backbone via a lightweight MLP projector and adopts an instruction-tuning pipeline inspired by LLaVA. By incorporating dense human-authored descriptions from the DOCCI dataset, VisionPangu improves semantic coherence and descriptive richness without relying on aggressive model scaling. Experimental results demonstrate that compact multimodal models can achieve competitive performance while producing more structured and detailed captions. The code and model weights will be publicly available at https://www.modelscope.cn/models/asdfgh007/visionpangu.		None
EDITOR: Effective and Interpretable Prompt Inversion for Text-to-Image Diffusion Models	2026-03-04	Show Text-to-image generation models~(e.g., Stable Diffusion) have achieved significant advancements, enabling the creation of high-quality and realistic images based on textual descriptions. Prompt inversion, the task of identifying the textual prompt used to generate a specific artifact, holds significant potential for applications including data attribution, model provenance, and watermarking validation. Recent studies introduced a delayed projection scheme to optimize for prompts representative of the vocabulary space, though challenges in semantic fluency and efficiency remain. Advanced image captioning models or visual large language models can generate highly interpretable prompts, but they often lack in image similarity. In this paper, we propose a prompt inversion technique called \sys for text-to-image diffusion models, which includes initializing embeddings using a pre-trained image captioning model, refining them through reverse-engineering in the latent space, and converting them to texts using an embedding-to-text model. Our experiments on the widely-used datasets, such as MS COCO, LAION, Flickr and DiffusionDB, show that our method outperforms existing methods in terms of image similarity, textual alignment, prompt interpretability and generalizability. We further illustrate the application of our generated prompts in tasks such as cross-concept image synthesis, concept manipulation, evolutionary multi-concept generation and unsupervised segmentation.		None
Cross-modal Identity Mapping: Minimizing Information Loss in Modality Conversion via Reinforcement Learning	2026-03-02	Show Large Vision-Language Models (LVLMs) often omit or misrepresent critical visual content in generated image captions. Minimizing such information loss will force LVLMs to focus on image details to generate precise descriptions. However, measuring information loss during modality conversion is inherently challenging due to the modal gap between visual content and text output. In this paper, we argue that the quality of an image caption is positively correlated with the similarity between images retrieved via text search using that caption. Based on this insight, we further propose Cross-modal Identity Mapping (CIM), a reinforcement learning framework that enhances image captioning without requiring additional annotations. Specifically, the method quantitatively evaluates the information loss from two perspectives: Gallery Representation Consistency and Query-gallery Image Relevance. Supervised under these metrics, LVLM minimizes information loss and aims to achieve identity mapping from images to captions. The experimental results demonstrate the superior performance of our method in image captioning, even when compared with Supervised Fine-Tuning. Particularly, on the COCO-LN500 benchmark, CIM achieves a 20% improvement in relation reasoning on Qwen2.5-VL-7B.The code will be released when the paper is accepted.	Accep... Accepted by CVPR 2026	None
MMCOMET: A Large-Scale Multimodal Commonsense Knowledge Graph for Contextual Reasoning	2026-03-01	Show We present MMCOMET, the first multimodal commonsense knowledge graph (MMKG) that integrates physical, social, and eventive knowledge. MMCOMET extends the ATOMIC2020 knowledge graph to include a visual dimension, through an efficient image retrieval process, resulting in over 900K multimodal triples. This new resource addresses a major limitation of existing MMKGs in supporting complex reasoning tasks like image captioning and storytelling. Through a standard visual storytelling experiment, we show that our holistic approach enables the generation of richer, coherent, and contextually grounded stories than those produced using text-only knowledge. This resource establishes a new foundation for multimodal commonsense reasoning and narrative generation.		None
CaptionFool: Universal Image Captioning Model Attacks	2026-02-28	Show Image captioning models are encoder-decoder architectures trained on large-scale image-text datasets, making them susceptible to adversarial attacks. We present CaptionFool, a novel universal (input-agnostic) adversarial attack against state-of-the-art transformer-based captioning models. By modifying only 7 out of 577 image patches (approximately 1.2% of the image), our attack achieves 94-96% success rate in generating arbitrary target captions, including offensive content. We further demonstrate that CaptionFool can generate "slang" terms specifically designed to evade existing content moderation filters. Our findings expose critical vulnerabilities in deployed vision-language models and underscore the urgent need for robust defenses against such attacks. Warning: This paper contains model outputs which are offensive in nature.		None
SesaHand: Enhancing 3D Hand Reconstruction via Controllable Generation with Semantic and Structural Alignment	2026-02-28	Show Recent studies on 3D hand reconstruction have demonstrated the effectiveness of synthetic training data to improve estimation performance. However, most methods rely on game engines to synthesize hand images, which often lack diversity in textures and environments, and fail to include crucial components like arms or interacting objects. Generative models are promising alternatives to generate diverse hand images, but still suffer from misalignment issues. In this paper, we present SesaHand, which enhances controllable hand image generation from both semantic and structural alignment perspectives for 3D hand reconstruction. Specifically, for semantic alignment, we propose a pipeline with Chain-of-Thought inference to extract human behavior semantics from image captions generated by the Vision-Language Model. This semantics suppresses human-irrelevant environmental details and ensures sufficient human-centric contexts for hand image generation. For structural alignment, we introduce hierarchical structural fusion to integrate structural information with different granularity for feature refinement to better align the hand and the overall human body in generated images. We further propose a hand structure attention enhancement method to efficiently enhance the model's attention on hand regions. Experiments demonstrate that our method not only outperforms prior work in generation performance but also improves 3D hand reconstruction with the generated hand images.		None
Hyperdimensional Cross-Modal Alignment of Frozen Language and Image Models for Efficient Image Captioning	2026-02-27	Show Large unimodal foundation models for vision and language encode rich semantic structures, yet aligning them typically requires computationally intensive multimodal fine-tuning. Such approaches depend on large-scale parameter updates, are resource intensive, and can perturb pretrained representations. Emerging evidence suggests, however, that independently trained foundation models may already exhibit latent semantic compatibility, reflecting shared structures in the data they model. This raises a fundamental question: can cross-modal alignment be achieved without modifying the models themselves? Here we introduce HDFLIM (HyperDimensional computing with Frozen Language and Image Models), a framework that establishes cross-modal mappings while keeping pretrained vision and language models fully frozen. HDFLIM projects unimodal embeddings into a shared hyperdimensional space and leverages lightweight symbolic operations -- binding, bundling, and similarity-based retrieval to construct associative cross-modal representations in a single pass over the data. Caption generation emerges from high-dimensional memory retrieval rather than iterative gradient-based optimization. We show that HDFLIM achieves performance comparable to end-to-end vision-language training methods and produces captions that are more semantically grounded than zero-shot baselines. By decoupling alignment from parameter tuning, our results suggest that semantic mapping across foundation models can be realized through symbolic operations on hyperdimensional encodings of the respective embeddings. More broadly, this work points toward an alternative paradigm for foundation model alignment in which frozen models are integrated through structured representational mappings rather than through large-scale retraining. The codebase for our implementation can be found at https://github.com/Abhishek-Dalvi410/HDFLIM.		Code Link
MovieTeller: Tool-augmented Movie Synopsis with ID Consistent Progressive Abstraction	2026-02-26	Show With the explosive growth of digital entertainment, automated video summarization has become indispensable for applications such as content indexing, personalized recommendation, and efficient media archiving. Automatic synopsis generation for long-form videos, such as movies and TV series, presents a significant challenge for existing Vision-Language Models (VLMs). While proficient at single-image captioning, these general-purpose models often exhibit critical failures in long-duration contexts, primarily a lack of ID-consistent character identification and a fractured narrative coherence. To overcome these limitations, we propose MovieTeller, a novel framework for generating movie synopses via tool-augmented progressive abstraction. Our core contribution is a training-free, tool-augmented, fact-grounded generation process. Instead of requiring costly model fine-tuning, our framework directly leverages off-the-shelf models in a plug-and-play manner. We first invoke a specialized face recognition model as an external "tool" to establish Factual Groundings--precise character identities and their corresponding bounding boxes. These groundings are then injected into the prompt to steer the VLM's reasoning, ensuring the generated scene descriptions are anchored to verifiable facts. Furthermore, our progressive abstraction pipeline decomposes the summarization of a full-length movie into a multi-stage process, effectively mitigating the context length limitations of current VLMs. Experiments demonstrate that our approach yields significant improvements in factual accuracy, character consistency, and overall narrative coherence compared to end-to-end baselines.	6 pages, CSCWD 2026	None
CLIP-Free, Label Free, Unsupervised Concept Bottleneck Models	2026-02-26	Show Concept Bottleneck Models (CBMs) map dense feature representations into human-interpretable concepts which are then combined linearly to make a prediction. However, modern CBMs rely on the CLIP model to obtain image-concept annotations, and it remains unclear how to design CBMs without the CLIP bottleneck. Methods that do not use CLIP instead require manual, labor intensive annotation to associate feature representations with concepts. Furthermore, all CBMs necessitate training a linear classifier to map the extracted concepts to class labels. In this work, we lift all three limitations simultaneously by proposing a method that converts any frozen visual classifier into a CBM without requiring image-concept labels (label-free), without relying on the CLIP model (CLIP-free), and by deriving the linear classifier in an unsupervised manner. Our method is formulated by aligning the original classifier's distribution (over discrete class indices) with its corresponding vision-language counterpart distribution derived from textual class names, while preserving the classifier's performance. The approach requires no ground-truth image-class annotations, and is highly data-efficient and preserves the classifier's reasoning process. Applied and tested on over 40 visual classifiers, our resulting unsupervised, label-free and CLIP-free CBM (U-F$^2$-CBM) sets a new state of the art, surpassing even supervised CLIP-based CBMs. We also show that our method can be used for zero-shot image captioning, outperforming existing methods based on CLIP, and achieving state-of-art.	CVPR 2026 (Findings)	None
LLM2CLIP: Powerful Language Model Unlocks Richer Cross-Modality Representation	2026-02-25	Show CLIP is a seminal multimodal model that maps images and text into a shared representation space through contrastive learning on billions of image-caption pairs. Inspired by the rapid progress of large language models (LLMs), we investigate how the superior linguistic understanding and broad world knowledge of LLMs can further strengthen CLIP, particularly in handling long and complex captions. We introduce an efficient fine-tuning framework that embeds an LLM into a pretrained CLIP while incurring nearly the same training cost as standard CLIP fine-tuning. Our method first converts the LLM into an embedding-compatible form for the CLIP setting, and then couples it with the pretrained CLIP vision encoder through a lightweight adaptor trained on only a few million image-caption pairs. With this strategy, we achieve large performance gains without large-scale retraining, outperforming state-of-the-art CLIP variants such as EVA02 and SigLIP-2. The LLM-enhanced CLIP delivers consistent improvements across a wide range of downstream tasks, including linear-probe classification, zero-shot image-text retrieval with both short and long captions (in English and other languages), zero-shot and supervised image segmentation, object detection, and serving as a tokenizer backbone for multimodal large-model benchmarks. Code and models are available at: https://aka.ms/llm2clip		None
CCCaption: Dual-Reward Reinforcement Learning for Complete and Correct Image Captioning	2026-02-25	Show Image captioning remains a fundamental task for vision language understanding, yet ground-truth supervision still relies predominantly on human-annotated references. Because human annotations reflect subjective preferences and expertise, ground-truth captions are often incomplete or even incorrect, which in turn limits caption models. We argue that caption quality should be assessed by two objective aspects: completeness (does the caption cover all salient visual facts?) and correctness (are the descriptions true with respect to the image?). To this end, we introduce CCCaption: a dual-reward reinforcement learning framework with a dedicated fine-tuning corpus that explicitly optimizes these properties to generate \textbf{C}omplete and \textbf{C}orrect \textbf{Captions}. For completeness, we use diverse LVLMs to disentangle the image into a set of visual queries, and reward captions that answer more of these queries, with a dynamic query sampling strategy to improve training efficiency. For correctness, we penalize captions that contain hallucinations by validating the authenticity of sub-caption queries, which are derived from the caption decomposition. Our symmetric dual-reward optimization jointly maximizes completeness and correctness, guiding models toward captions that better satisfy these objective criteria. Extensive experiments across standard captioning benchmarks show consistent improvements, offering a principled path to training caption models beyond human-annotation imitation.	Accept by CVPR 2026	None
Closing the gap in multimodal medical representation alignment	2026-02-23	Show In multimodal learning, CLIP has emerged as the de-facto approach for mapping different modalities into a shared latent space by bringing semantically similar representations closer while pushing apart dissimilar ones. However, CLIP-based contrastive losses exhibit unintended behaviors that negatively impact true semantic alignment, leading to sparse and fragmented latent spaces. This phenomenon, known as the modality gap, has been partially mitigated for standard text and image pairs but remains unknown and unresolved in more complex multimodal settings, such as the medical domain. In this work, we study this phenomenon in the latter case, revealing that the modality gap is present also in medical alignment, and we propose a modality-agnostic framework that closes this gap, ensuring that semantically related representations are more aligned, regardless of their source modality. Our method enhances alignment between radiology images and clinical text, improving cross-modal retrieval and image captioning.	Accepted at MLSP2025	None
Knowledge-aware Visual Question Generation for Remote Sensing Images	2026-02-22	Show With the rapid development of remote sensing image archives, asking questions about images has become an effective way of gathering specific information or performing image retrieval. However, automatically generated image-based questions tend to be simplistic and template-based, which hinders the real deployment of question answering or visual dialogue systems. To enrich and diversify the questions, we propose a knowledge-aware remote sensing visual question generation model, KRSVQG, that incorporates external knowledge related to the image content to improve the quality and contextual understanding of the generated questions. The model takes an image and a related knowledge triplet from external knowledge sources as inputs and leverages image captioning as an intermediary representation to enhance the image grounding of the generated questions. To assess the performance of KRSVQG, we utilized two datasets that we manually annotated: NWPU-300 and TextRS-300. Results on these two datasets demonstrate that KRSVQG outperforms existing methods and leads to knowledge-enriched questions, grounded in both image and domain knowledge.		None
Questions beyond Pixels: Integrating Commonsense Knowledge in Visual Question Generation for Remote Sensing	2026-02-22	Show With the rapid development of remote sensing image archives, asking questions about images has become an effective way of gathering specific information or performing semantic image retrieval. However, current automatically generated questions tend to be simplistic and template-based, which hinders the deployment of question answering or visual dialogue systems for real-world applications. To enrich and diversify the questions with both image content and commonsense knowledge, we propose a Knowledge-aware Remote Sensing Visual Question Generation model (KRSVQG). The proposed model incorporates related knowledge triplets from external knowledge sources to broaden the question content, while employing image captioning as an intermediary representation to ground questions to the corresponding images. Moreover, KRSVQG utilizes a vision-language pre-training and fine-tuning strategy, enabling the model's adaptation to low data regimes. To evaluate the proposed KRSVQG model, we construct two knowledge-aware remote sensing visual question generation datasets: the NWPU-300 dataset and the TextRS-300 dataset. Evaluations, including metrics and human assessment, demonstrate that KRSVQG outperforms existing methods and leads to rich questions, grounded in both image and domain knowledge. As a key practice in vision-language research, knowledge-aware visual question generation advances the understanding of image content beyond pixels, facilitating the development of knowledge-enriched vision-language systems with vision-grounded human commonsense.		None
Initialization matters in few-shot adaptation of vision-language models for histopathological image classification	2026-02-21	Show Vision language models (VLM) pre-trained on datasets of histopathological image-caption pairs enabled zero-shot slide-level classification. The ability of VLM image encoders to extract discriminative features also opens the door for supervised fine-tuning for whole-slide image (WSI) classification, ideally using few labeled samples. Slide-level prediction frameworks require the incorporation of multiple instance learning (MIL) due to the gigapixel size of the WSI. Following patch-level feature extraction and aggregation, MIL frameworks rely on linear classifiers trained on top of the slide-level aggregated features. Classifier weight initialization has a large influence on Linear Probing performance in efficient transfer learning (ETL) approaches based on few-shot learning. In this work, we propose Zero-Shot Multiple-Instance Learning (ZS-MIL) to address the limitations of random classifier initialization that underperform zero-shot prediction in MIL problems. ZS-MIL uses the class-level embeddings of the VLM text encoder as the classification layer's starting point to compute each sample's bag-level probabilities. Through multiple experiments, we demonstrate the robustness of ZS-MIL compared to well-known weight initialization techniques both in terms of performance and variability in an ETL few-shot scenario for subtyping prediction.	Accep... Accepted as oral presentation at CASEIB 2024 held in Sevilla, Spain	None
MiSCHiEF: A Benchmark in Minimal-Pairs of Safety and Culture for Holistic Evaluation of Fine-Grained Image-Caption Alignment	2026-02-21	Show Fine-grained image-caption alignment is crucial for vision-language models (VLMs), especially in socially critical contexts such as identifying real-world risk scenarios or distinguishing cultural proxies, where correct interpretation hinges on subtle visual or linguistic clues and where minor misinterpretations can lead to significant real-world consequences. We present MiSCHiEF, a set of two benchmarking datasets based on a contrastive pair design in the domains of safety (MiS) and culture (MiC), and evaluate four VLMs on tasks requiring fine-grained differentiation of paired images and captions. In both datasets, each sample contains two minimally differing captions and corresponding minimally differing images. In MiS, the image-caption pairs depict a safe and an unsafe scenario, while in MiC, they depict cultural proxies in two distinct cultural contexts. We find that models generally perform better at confirming the correct image-caption pair than rejecting incorrect ones. Additionally, models achieve higher accuracy when selecting the correct caption from two highly similar captions for a given image, compared to the converse task. The results, overall, highlight persistent modality misalignment challenges in current VLMs, underscoring the difficulty of precise cross-modal grounding required for applications with subtle semantic and visual distinctions.	EACL ... EACL 2026, Main, Short Paper	None
EarthSpatialBench: Benchmarking Spatial Reasoning Capabilities of Multimodal LLMs on Earth Imagery	2026-02-17	Show Benchmarking spatial reasoning in multimodal large language models (MLLMs) has attracted growing interest in computer vision due to its importance for embodied AI and other agentic systems that require precise interaction with the physical world. However, spatial reasoning on Earth imagery has lagged behind, as it uniquely involves grounding objects in georeferenced images and quantitatively reasoning about distances, directions, and topological relations using both visual cues and vector geometry coordinates (e.g., 2D bounding boxes, polylines, and polygons). Existing benchmarks for Earth imagery primarily focus on 2D spatial grounding, image captioning, and coarse spatial relations (e.g., simple directional or proximity cues). They lack support for quantitative direction and distance reasoning, systematic topological relations, and complex object geometries beyond bounding boxes. To fill this gap, we propose \textbf{EarthSpatialBench}, a comprehensive benchmark for evaluating spatial reasoning in MLLMs on Earth imagery. The benchmark contains over 325K question-answer pairs spanning: (1) qualitative and quantitative reasoning about spatial distance and direction; (2) systematic topological relations; (3) single-object queries, object-pair queries, and compositional aggregate group queries; and (4) object references expressed via textual descriptions, visual overlays, and explicit geometry coordinates, including 2D bounding boxes, polylines, and polygons. We conducted extensive experiments on both open-source and proprietary models to identify limitations in the spatial reasoning of MLLMs.		None
GMAIL: Generative Modality Alignment for generated Image Learning	2026-02-17	Show Generative models have made it possible to synthesize highly realistic images, potentially providing an abundant data source for training machine learning models. Despite the advantages of these synthesizable data sources, the indiscriminate use of generated images as real images for training can even cause mode collapse due to modality discrepancies between real and synthetic domains. In this paper, we propose a novel framework for discriminative use of generated images, coined GMAIL, that explicitly treats generated images as a separate modality from real images. Instead of indiscriminately replacing real images with generated ones in the pixel space, our approach bridges the two distinct modalities in the same latent space through a multi-modal learning approach. To be specific, we first fine-tune a model exclusively on generated images using a cross-modality alignment loss and then employ this aligned model to further train various vision-language models with generated images. By aligning the two modalities, our approach effectively leverages the benefits of recent advances in generative models, thereby boosting the effectiveness of generated image learning across a range of vision-language tasks. Our framework can be easily incorporated with various vision-language models, and we demonstrate its efficacy throughout extensive experiments. For example, our framework significantly improves performance on image captioning, zero-shot image retrieval, zero-shot image classification, and long caption retrieval tasks. It also shows positive generated data scaling trends and notable enhancements in the captioning performance of the large multimodal model, LLaVA.		None
Web-Scale Multimodal Summarization using CLIP-Based Semantic Alignment	2026-02-16	Show We introduce Web-Scale Multimodal Summarization, a lightweight framework for generating summaries by combining retrieved text and image data from web sources. Given a user-defined topic, the system performs parallel web, news, and image searches. Retrieved images are ranked using a fine-tuned CLIP model to measure semantic alignment with topic and text. Optional BLIP captioning enables image-only summaries for stronger multimodal coherence.The pipeline supports features such as adjustable fetch limits, semantic filtering, summary styling, and downloading structured outputs. We expose the system via a Gradio-based API with controllable parameters and preconfigured presets.Evaluation on 500 image-caption pairs with 20:1 contrastive negatives yields a ROC-AUC of 0.9270, an F1-score of 0.6504, and an accuracy of 96.99%, demonstrating strong multimodal alignment. This work provides a configurable, deployable tool for web-scale summarization that integrates language, retrieval, and vision models in a user-extensible pipeline.		None
Is Information Density Uniform when Utterances are Grounded on Perception and Discourse?	2026-02-16	Show The Uniform Information Density (UID) hypothesis posits that speakers are subject to a communicative pressure to distribute information evenly within utterances, minimising surprisal variance. While this hypothesis has been tested empirically, prior studies are limited exclusively to text-only inputs, abstracting away from the perceptual context in which utterances are produced. In this work, we present the first computational study of UID in visually grounded settings. We estimate surprisal using multilingual vision-and-language models over image-caption data in 30 languages and visual storytelling data in 13 languages, together spanning 11 families. We find that grounding on perception consistently smooths the distribution of information, increasing both global and local uniformity across typologically diverse languages compared to text-only settings. In visual narratives, grounding in both image and discourse contexts has additional effects, with the strongest surprisal reductions occurring at the onset of discourse units. Overall, this study takes a first step towards modelling the temporal dynamics of information flow in ecologically plausible, multimodal language use, and finds that grounded language exhibits greater information uniformity, supporting a context-sensitive formulation of UID.	Accep... Accepted as main paper at EACL 2026	None
Top-Down Semantic Refinement for Image Captioning	2026-02-16	Show Large Vision-Language Models (VLMs) face an inherent contradiction in image captioning: their powerful single-step generation capabilities often lead to a myopic decision-making process. This makes it difficult to maintain global narrative coherence while capturing rich details, a limitation that is particularly pronounced in tasks that require multi-step and complex scene description. To overcome this fundamental challenge, we redefine image captioning as a goal-oriented hierarchical refinement planning problem, and further propose a novel framework, named Top-Down Semantic Refinement (TDSR), which models the generation process as a Markov Decision Process (MDP). However, planning within the vast state space of a VLM presents a significant computational hurdle. Our core contribution, therefore, is the design of a highly efficient Monte Carlo Tree Search (MCTS) algorithm tailored for VLMs. By incorporating a visual-guided parallel expansion and a lightweight value network, our TDSR reduces the call frequency to the expensive VLM by an order of magnitude without sacrificing planning quality. Furthermore, an adaptive early stopping mechanism dynamically matches computational overhead to the image's complexity. Extensive experiments on multiple benchmarks, including DetailCaps, COMPOSITIONCAP, and POPE, demonstrate that our TDSR, as a plug-and-play module, can significantly enhance the performance of existing VLMs (e.g., LLaVA-1.5, Qwen2.5-VL) by achieving state-of-the-art or highly competitive results in fine-grained description, compositional generalization, and hallucination suppression.		None
RAVENEA: A Benchmark for Multimodal Retrieval-Augmented Visual Culture Understanding	2026-02-14	Show As vision-language models (VLMs) become increasingly integrated into daily life, the need for accurate visual culture understanding is becoming critical. Yet, these models frequently fall short in interpreting cultural nuances effectively. Prior work has demonstrated the effectiveness of retrieval-augmented generation (RAG) in enhancing cultural understanding in text-only settings, while its application in multimodal scenarios remains underexplored. To bridge this gap, we introduce RAVENEA (Retrieval-Augmented Visual culturE uNdErstAnding), a new benchmark designed to advance visual culture understanding through retrieval, focusing on two tasks: culture-focused visual question answering (cVQA) and culture-informed image captioning (cIC). RAVENEA extends existing datasets by integrating over 11,396 unique Wikipedia documents curated and ranked by human annotators. Through the extensive evaluation on seven multimodal retrievers and fifteen VLMs, RAVENEA reveals some undiscovered findings: (i) In general, cultural grounding annotations can enhance multimodal retrieval and corresponding downstream tasks. (ii) VLMs, when augmented with culture-aware retrieval, generally outperform their non-augmented counterparts (by averaging +6% on cVQA and +11% on cIC). (iii) Performance of culture-aware retrieval augmented varies widely across countries. These findings highlight the limitations of current multimodal retrievers and VLMs, underscoring the need to enhance visual culture understanding within RAG systems. We believe RAVENEA offers a valuable resource for advancing research on retrieval-augmented visual culture understanding.	ICLR ... ICLR 2026; Project page: https://jiaangli.github.io/ravenea/	Code Link
OmniScience: A Large-scale Multi-modal Dataset for Scientific Image Understanding	2026-02-14	Show Multimodal Large Language Models demonstrate strong performance on natural image understanding, yet exhibit limited capability in interpreting scientific images, including but not limited to schematic diagrams, experimental characterizations, and analytical charts. This limitation is particularly pronounced in open-source MLLMs. The gap largely stems from existing datasets with limited domain coverage, coarse structural annotations, and weak semantic grounding. We introduce OmniScience, a large-scale, high-fidelity multi-modal dataset comprising 1.5 million figure-caption-context triplets, spanning more than 10 major scientific disciplines. To obtain image caption data with higher information density and accuracy for multi-modal large-model training, we develop a dynamic model-routing re-captioning pipeline that leverages state-of-the-art multi-modal large language models to generate dense, self-contained descriptions by jointly synthesizing visual features, original figure captions, and corresponding in-text references authored by human scientists. The pipeline is further reinforced with rigorous quality filtering and alignment with human expert judgments, ensuring both factual accuracy and semantic completeness, and boosts the image-text multi-modal similarity score from 0.769 to 0.956. We further propose a caption QA protocol as a proxy task for evaluating visual understanding. Under this setting, Qwen2.5-VL-3B model finetuned on OmniScience show substantial gains over baselines, achieving a gain of 0.378 on MM-MT-Bench and a gain of 0.140 on MMMU.		None
MRMR: A Realistic and Expert-Level Multidisciplinary Benchmark for Reasoning-Intensive Multimodal Retrieval	2026-02-14	Show We introduce MRMR, the first expert-level multidisciplinary multimodal retrieval benchmark requiring intensive reasoning. MRMR contains 1,502 queries spanning 23 domains, with positive documents carefully verified by human experts. Compared to prior benchmarks, MRMR introduces three key advancements. First, it challenges retrieval systems across diverse areas of expertise, enabling fine-grained model comparison across domains. Second, queries are reasoning-intensive, with images requiring deeper interpretation such as diagnosing microscopic slides. We further introduce Contradiction Retrieval, a novel task requiring models to identify conflicting concepts. Finally, queries and documents are constructed as image-text interleaved sequences. Unlike earlier benchmarks restricted to single images or unimodal documents, MRMR offers a realistic setting with multi-image queries and mixed-modality corpus documents. We conduct an extensive evaluation of 4 categories of multimodal retrieval systems and 14 frontier models on MRMR. The text embedding model Qwen3-Embedding with LLM-generated image captions achieves the highest performance, highlighting substantial room for improving multimodal retrieval models. Although latest multimodal models such as Ops-MM-Embedding perform competitively on expert-domain queries, they fall short on reasoning-intensive tasks. We believe that MRMR paves the way for advancing multimodal retrieval in more realistic and challenging scenarios.		None
Using Deep Learning to Generate Semantically Correct Hindi Captions	2026-02-13	Show Automated image captioning using the content from the image is very appealing when done by harnessing the capability of computer vision and natural language processing. Extensive research has been done in the field with a major focus on the English language which gives the scope for further developments in the same with consideration of popular foreign languages. This research utilizes distinct models for translating the image caption into Hindi, the fourth most popular language across the world. Exploring the multi-modal architectures this research comprises local visual features, global visual features, attention mechanisms, and pre-trained models. Using google cloud translator on the image dataset from Flickr8k, Hindi image descriptions have been generated. Pre-trained CNNs like VGG16, ResNet50, and Inception V3 helped in retrieving image characteristics, while the uni-directional and bi-directional techniques of text encoding are used for the text encoding process. An additional Attention layer helps to generate a weight vector and, by multiplying it, combine image characteristics from each time step into a sentence-level feature vector. Bilingual evaluation understudy scores are used to compare the research outcome. Many experiments that serve as a baseline are done for the comparative analysis of the research. An image with a score of BLEU-1 is considered sufficient, whereas one with a score of BLEU-4 is considered to have fluid image captioning. For both BLEU scores, the attention-based bidirectional LSTM with VGG16 produced the best results of 0.59 and 0.19 respectively. The experiments conclude that researchs ability to produce relevant, semantically accurate image captions in Hindi. The research accomplishes the goals and future research can be guided by this research model.	34 pa... 34 pages, 12 figures, 3 tables. Master's thesis, Liverpool John Moores University, November 2022	None
Remote Sensing Retrieval-Augmented Generation: Bridging Remote Sensing Imagery and Comprehensive Knowledge with a Multi-Modal Dataset and Retrieval-Augmented Generation Model	2026-02-12	Show Recent progress in VLMs has demonstrated impressive capabilities across a variety of tasks in the natural image domain. Motivated by these advancements, the remote sensing community has begun to adopt VLMs for remote sensing vision-language tasks, including scene understanding, image captioning, and visual question answering. However, existing remote sensing VLMs typically rely on closed-set scene understanding and focus on generic scene descriptions, yet lack the ability to incorporate external knowledge. This limitation hinders their capacity for semantic reasoning over complex or context-dependent queries that involve domain-specific or world knowledge. To address these challenges, we first introduced a multimodal Remote Sensing World Knowledge (RSWK) dataset, which comprises high-resolution satellite imagery and detailed textual descriptions for 14,141 well-known landmarks from 175 countries, integrating both remote sensing domain knowledge and broader world knowledge. Building upon this dataset, we proposed a novel Remote Sensing Retrieval-Augmented Generation (RS-RAG) framework, which consists of two key components. The Multi-Modal Knowledge Vector Database Construction module encodes remote sensing imagery and associated textual knowledge into a unified vector space. The Knowledge Retrieval and Response Generation module retrieves and re-ranks relevant knowledge based on image and/or text queries, and incorporates the retrieved content into a knowledge-augmented prompt to guide the VLM in producing contextually grounded responses. We validated the effectiveness of our approach on three representative vision-language tasks, including image captioning, image classification, and visual question answering, where RS-RAG significantly outperformed state-of-the-art baselines.	Accep... Accepted by IEEE Geoscience and Remote Sensing Magazine (GRSM)	None
Where Do Images Come From? Analyzing Captions to Geographically Profile Datasets	2026-02-10	Show Recent studies show that text-to-image models often fail to generate geographically representative images, raising concerns about the representativeness of their training data and motivating the question: which parts of the world do these training examples come from? We geographically profile large-scale multimodal datasets by mapping image-caption pairs to countries based on location information extracted from captions using LLMs. Studying English captions from three widely used datasets (Re-LAION, DataComp1B, and Conceptual Captions) across $20$ common entities (e.g., house, flag), we find that the United States, the United Kingdom, and Canada account for $48.0%$ of samples, while South American and African countries are severely under-represented with only $1.8%$ and $3.8%$ of images, respectively. We observe a strong correlation between a country's GDP and its representation in the data ($ρ= 0.82$). Examining non-English subsets for $4$ languages from the Re-LAION dataset, we find that representation skews heavily toward countries where these languages are predominantly spoken. Additionally, we find that higher representation does not necessarily translate to greater visual or semantic diversity. Finally, analyzing country-specific images generated by Stable Diffusion v1.3 trained on Re-LAION, we show that while generations appear realistic, they are severely limited in their coverage compared to real-world images.	41 pages, 20 figures	None
MILE-RefHumEval: A Reference-Free, Multi-Independent LLM Framework for Human-Aligned Evaluation	2026-02-10	Show We introduce MILE-RefHumEval, a reference-free framework for evaluating Large Language Models (LLMs) without ground-truth annotations or evaluator coordination. It leverages an ensemble of independently prompted evaluators guided by a human-aligned schema, supporting both discrete and continuous scoring judgement. With task-specific prompts from best candidate selection, summarization and image captioning to dialogue, MILE-RefHumEval provides flexible, interpretable, and scalable assessments. Experiments show it aligns closely with human judgments, outperforms prior methods, and reduces computational overhead, offering an efficient, robust, and human-aligned solution for real-world LLM evaluation.		None
Robust Pre-Training of Medical Vision-and-Language Models with Domain-Invariant Multi-Modal Masked Reconstruction	2026-02-06	Show Medical vision-language models show strong potential for joint reasoning over medical images and clinical text, but their performance often degrades under domain shift caused by variations in imaging devices, acquisition protocols, and reporting styles. Existing multi-modal pre-training methods largely overlook robustness, treating it as a downstream adaptation problem. In this work, we propose Robust Multi-Modal Masked Reconstruction (Robust-MMR), a self-supervised pre-training framework that explicitly incorporates robustness objectives into masked vision-language learning. Robust-MMR integrates asymmetric perturbation-aware masking, domain-consistency regularization, and modality-resilience constraints to encourage domain-invariant representations. We evaluate Robust-MMR on multiple medical vision-language benchmarks, including medical visual question answering (VQA-RAD, SLAKE, VQA-2019), cross-domain image-text classification (MELINDA), and robust image-caption retrieval (ROCO). Robust-MMR achieves 78.9% cross-domain accuracy on VQA-RAD, outperforming the strongest baseline by 3.8 percentage points, and reaches 74.6% and 77.0% accuracy on SLAKE and VQA-2019, respectively. Under perturbed evaluation, Robust-MMR improves VQA-RAD accuracy from 69.1% to 75.6%. For image-text classification, cross-domain MELINDA accuracy increases from 70.3% to 75.2%, while retrieval experiments show a reduction in mean rank degradation from over 16 to 4.1 under perturbation. Qualitative results further demonstrate improved clinical reasoning for disease detection and structural abnormality assessment. These findings show that explicitly modeling robustness during pre-training leads to more reliable and transferable medical vision-language representations for real-world deployment.	28 pages, 3 figures	None
PhenoLIP: Integrating Phenotype Ontology Knowledge into Medical Vision-Language Pretraining	2026-02-05	Show Recent progress in large-scale CLIP-like vision-language models(VLMs) has greatly advanced medical image analysis. However, most existing medical VLMs still rely on coarse image-text contrastive objectives and fail to capture the systematic visual knowledge encoded in well-defined medical phenotype ontologies. To address this gap, we construct PhenoKG, the first large-scale, phenotype-centric multimodal knowledge graph that encompasses over 520K high-quality image-text pairs linked to more than 3,000 phenotypes. Building upon PhenoKG, we propose PhenoLIP, a novel pretraining framework that explicitly incorporates structured phenotype knowledge into medical VLMs through a two-stage process. We first learn a knowledge-enhanced phenotype embedding space from textual ontology data and then distill this structured knowledge into multimodal pretraining via a teacher-guided knowledge distillation objective. To support evaluation, we further introduce PhenoBench, an expert-verified benchmark designed for phenotype recognition, comprising over 7,800 image--caption pairs covering more than 1,000 phenotypes. Extensive experiments demonstrate that PhenoLIP outperforms previous state-of-the-art baselines, improving upon BiomedCLIP in phenotype classification accuracy by 8.85% and BIOMEDICA in cross-modal retrieval by 15.03%, underscoring the value of integrating phenotype-centric priors into medical VLMs for structured and interpretable medical image understanding.		None
CLIP-Map: Structured Matrix Mapping for Parameter-Efficient CLIP Compression	2026-02-05	Show Contrastive Language-Image Pre-training (CLIP) has achieved widely applications in various computer vision tasks, e.g., text-to-image generation, Image-Text retrieval and Image captioning. However, CLIP suffers from high memory and computation cost, which prohibits its usage to the resource-limited application scenarios. Existing CLIP compression methods typically reduce the size of pre-trained CLIP weights by selecting their subset as weight inheritance for further retraining via mask optimization or important weight measurement. However, these select-based weight inheritance often compromises the feature presentation ability, especially on the extreme compression. In this paper, we propose a novel mapping-based CLIP compression framework, CLIP-Map. It leverages learnable matrices to map and combine pretrained weights by Full-Mapping with Kronecker Factorization, aiming to preserve as much information from the original weights as possible. To mitigate the optimization challenges introduced by the learnable mapping, we propose Diagonal Inheritance Initialization to reduce the distribution shifting problem for efficient and effective mapping learning. Extensive experimental results demonstrate that the proposed CLIP-Map outperforms select-based frameworks across various compression ratios, with particularly significant gains observed under high compression settings.		None
VEAttack: Downstream-agnostic Vision Encoder Attack against Large Vision Language Models	2026-02-04	Show Large Vision-Language Models (LVLMs) have demonstrated remarkable capabilities in multimodal understanding and generation, yet their vulnerability to adversarial attacks raises significant robustness concerns. While existing effective attacks always focus on task-specific white-box settings, these approaches are limited in the context of LVLMs, which are designed for diverse downstream tasks and require expensive full-model gradient computations. Motivated by the pivotal role and wide adoption of the vision encoder in LVLMs, we propose a simple yet effective Vision Encoder Attack (VEAttack), which targets the vision encoder of LVLMs only. Specifically, we propose to generate adversarial examples by minimizing the cosine similarity between the clean and perturbed visual features, without accessing the following large language models, task information, and labels. It significantly reduces the computational overhead while eliminating the task and label dependence of traditional white-box attacks in LVLMs. To make this simple attack effective, we propose to perturb images by optimizing image tokens instead of the classification token. We provide both empirical and theoretical evidence that VEAttack can easily generalize to various tasks. VEAttack has achieved a performance degradation of 94.5% on image caption task and 75.7% on visual question answering task. We also reveal some key observations to provide insights into LVLM attack/defense: 1) hidden layer variations of LLM, 2) token attention differential, 3) Möbius band in transfer attack, 4) low sensitivity to attack steps. The code is available at https://github.com/hefeimei06/VEAttack-LVLM.		Code Link
PromptSplit: Revealing Prompt-Level Disagreement in Generative Models	2026-02-03	Show Prompt-guided generative AI models have rapidly expanded across vision and language domains, producing realistic and diverse outputs from textual inputs. The growing variety of such models, trained with different data and architectures, calls for principled methods to identify which types of prompts lead to distinct model behaviors. In this work, we propose PromptSplit, a kernel-based framework for detecting and analyzing prompt-dependent disagreement between generative models. For each compared model pair, PromptSplit constructs a joint prompt--output representation by forming tensor-product embeddings of the prompt and image (or text) features, and then computes the corresponding kernel covariance matrix. We utilize the eigenspace of the weighted difference between these matrices to identify the main directions of behavioral difference across prompts. To ensure scalability, we employ a random-projection approximation that reduces computational complexity to $O(nr^2 + r^3)$ for projection dimension $r$. We further provide a theoretical analysis showing that this approximation yields an eigenstructure estimate whose expected deviation from the full-dimensional result is bounded by $O(1/r^2)$. Experiments across text-to-image, text-to-text, and image-captioning settings demonstrate that PromptSplit accurately detects ground-truth behavioral differences and isolates the prompts responsible, offering an interpretable tool for detecting where generative models disagree.		None
Contextualized Visual Personalization in Vision-Language Models	2026-02-03	Show Despite recent progress in vision-language models (VLMs), existing approaches often fail to generate personalized responses based on the user's specific experiences, as they lack the ability to associate visual inputs with a user's accumulated visual-textual context. We newly formalize this challenge as contextualized visual personalization, which requires the visual recognition and textual retrieval of personalized visual experiences by VLMs when interpreting new images. To address this issue, we propose CoViP, a unified framework that treats personalized image captioning as a core task for contextualized visual personalization and improves this capability through reinforcement-learning-based post-training and caption-augmented generation. We further introduce diagnostic evaluations that explicitly rule out textual shortcut solutions and verify whether VLMs truly leverage visual context. Extensive experiments demonstrate that existing open-source and proprietary VLMs exhibit substantial limitations, while CoViP not only improves personalized image captioning but also yields holistic gains across downstream personalization tasks. These results highlight CoViP as a crucial stage for enabling robust and generalizable contextualized visual personalization.	Proje... Project Page: https://github.com/oyt9306/CoViP	Code Link
Generative Engine Optimization: A VLM and Agent Framework for Pinterest Acquisition Growth	2026-02-03	Show Large Language Models are fundamentally reshaping content discovery through AI-native search systems such as ChatGPT, Gemini, and Claude. Unlike traditional search engines that match keywords to documents, these systems infer user intent, synthesize multimodal evidence, and generate contextual answers directly on the search page, introducing a paradigm shift from Search Engine Optimization (SEO) to Generative Engine Optimization (GEO). For visual content platforms hosting billions of assets, this poses an acute challenge: individual images lack the semantic depth and authority signals that generative search prioritizes, risking disintermediation as user needs are satisfied in-place without site visits. We present Pinterest GEO, a production-scale framework that pioneers reverse search design: rather than generating generic image captions describing what content is, we fine-tune Vision-Language Models (VLMs) to predict what users would actually search for, augmented this with AI agents that mine real-time internet trends to capture emerging search demand. These VLM-generated queries then drive construction of semantically coherent Collection Pages via multimodal embeddings, creating indexable aggregations optimized for generative retrieval. Finally, we employ hybrid VLM and two-tower ANN architectures to build authority-aware interlinking structures that propagate signals across billions of visual assets. Deployed at scale across billions of images and tens of millions of collections, GEO delivers 20% organic traffic growth contributing to multi-million monthly active user (MAU) growth, demonstrating a principled pathway for visual platforms to thrive in the generative search era.		None
DoubleTake: Contrastive Reasoning for Faithful Decision-Making in Medical Imaging	2026-02-02	Show Accurate decision making in medical imaging requires reasoning over subtle visual differences between confusable conditions, yet most existing approaches rely on nearest neighbor retrieval that returns redundant evidence and reinforces a single hypothesis. We introduce a contrastive, document-aware reference selection framework that constructs compact evidence sets optimized for discrimination rather than similarity by explicitly balancing visual relevance, embedding diversity, and source-level provenance using ROCO embeddings and metadata. While ROCO provides large-scale image-caption pairs, it does not specify how references should be selected for contrastive reasoning, and naive retrieval frequently yields near-duplicate figures from the same document. To address this gap, we release a reproducible reference selection protocol and curated reference bank that enable a systematic study of contrastive retrieval in medical image reasoning. Building on these contrastive evidence sets, we propose Counterfactual-Contrastive Inference, a confidence-aware reasoning framework that performs structured pairwise visual comparisons and aggregates evidence using margin-based decision rules with faithful abstention. On the MediConfusion benchmark, our approach achieves state-of-the-art performance, improving set-level accuracy by nearly 15% relative to prior methods while reducing confusion and improving individual accuracy.		None
Beyond Vision: Contextually Enriched Image Captioning with Multi-Modal Retrieval	2026-02-01	Show Real-world image captions often lack contextual depth, omitting crucial details such as event background, temporal cues, outcomes, and named entities that are not visually discernible. This gap limits the effectiveness of image understanding in domains like journalism, education, and digital archives, where richer, more informative descriptions are essential. To address this, we propose a multimodal pipeline that augments visual input with external textual knowledge. Our system retrieves semantically similar images using BEIT-3 (Flickr30k-384 and COCO-384) and SigLIP So-384, reranks them using ORB and SIFT for geometric alignment, and extracts contextual information from related articles via semantic search. A fine-tuned Qwen3 model with QLoRA then integrates this context with base captions generated by Instruct BLIP (Vicuna-7B) to produce event-enriched, context-aware descriptions. Evaluated on the OpenEvents v1 dataset, our approach generates significantly more informative captions compared to traditional methods, showing strong potential for real-world applications requiring deeper visual-textual understanding	7 pag... 7 pages, 5 figures. System description for the EVENTA Grand Challenge (Track 1) at ACM MM'25	None
HueManity: Probing Fine-Grained Visual Perception in MLLMs	2026-01-31	Show Recent Multimodal Large Language Models (MLLMs) demonstrate strong high-level visual reasoning on tasks such as visual question answering and image captioning. Yet existing benchmarks largely overlook their ability to capture fine-grained perceptual details. As MLLMs are increasingly deployed in safety and reliability critical settings, perceptual acuity becomes essential. We present HueManity, a scalable automated benchmark for assessing fine-grained visual perception in MLLMs. HueManity comprises 83,850 Ishihara-style images embedding alphanumeric strings, designed to evaluate pattern recognition, a core aspect of visual understanding. Our evaluation of nine state-of-the-art MLLMs uncovers a striking performance deficit: the strongest model achieved only 33.6% accuracy on a simple numeric task and 3% on a harder alphanumeric task, compared to near-ceiling performance from humans (99.38%, 93.25%) and a fine-tuned ResNet-50 (96.5%, 94.5%). These findings expose a critical weakness in MLLMs' perceptual grounding, one that remains obscured by conventional benchmarks emphasizing high-level semantics.		None
Brazilian Portuguese Image Captioning with Transformers: A Study on Cross-Native-Translated Dataset	2026-01-30	Show Image captioning (IC) refers to the automatic generation of natural language descriptions for images, with applications ranging from social media content generation to assisting individuals with visual impairments. While most research has been focused on English-based models, low-resource languages such as Brazilian Portuguese face significant challenges due to the lack of specialized datasets and models. Several studies create datasets by automatically translating existing ones to mitigate resource scarcity. This work addresses this gap by proposing a cross-native-translated evaluation of Transformer-based vision and language models for Brazilian Portuguese IC. We use a version of Flickr30K comprised of captions manually created by native Brazilian Portuguese speakers and compare it to a version with captions automatically translated from English to Portuguese. The experiments include a cross-context approach, where models trained on one dataset are tested on the other to assess the translation impact. Additionally, we incorporate attention maps for model inference interpretation and use the CLIP-Score metric to evaluate the image-description alignment. Our findings show that Swin-DistilBERTimbau consistently outperforms other models, demonstrating strong generalization across datasets. ViTucano, a Brazilian Portuguese pre-trained VLM, surpasses larger multilingual models (GPT-4o, LLaMa 3.2 Vision) in traditional text-based evaluation metrics, while GPT-4 models achieve the highest CLIP-Score, highlighting improved image-text alignment. Attention analysis reveals systematic biases, including gender misclassification, object enumeration errors, and spatial inconsistencies. The datasets and the models generated and analyzed during the current study are available in: https://github.com/laicsiifes/transformer-caption-ptbr.	Accep... Accepted to JBCS. 18 pages, 11 figures	Code Link
Modeling Image-Caption Rating from Comparative Judgments	2026-01-30	Show Rating the accuracy of captions in describing images is time-consuming and subjective for humans. In contrast, it is often easier for people to compare two captions and decide which one better matches a given image. In this work, we propose a machine learning framework that models such comparative judgments instead of direct ratings. The model can then be applied to rank unseen image-caption pairs in the same way as a regression model trained on direct ratings. Using the VICR dataset, we extract visual features with ResNet-50 and text features with MiniLM, then train both a regression model and a comparative learning model. While the regression model achieves better performance (Pearson's $ρ$: 0.7609 and Spearman's $r_s$: 0.7089), the comparative learning model steadily improves with more data and approaches the regression baseline. In addition, a small-scale human evaluation study comparing absolute rating, pairwise comparison, and same-image comparison shows that comparative annotation yields faster results and has greater agreement among human annotators. These results suggest that comparative learning can effectively model human preferences while significantly reducing the cost of human annotations.		None
Leveraging Data to Say No: Memory Augmented Plug-and-Play Selective Prediction	2026-01-30	Show Selective prediction aims to endow predictors with a reject option, to avoid low confidence predictions. However, existing literature has primarily focused on closed-set tasks, such as visual question answering with predefined options or fixed-category classification. This paper considers selective prediction for visual language foundation models, addressing a taxonomy of tasks ranging from closed to open set and from finite to unbounded vocabularies, as in image captioning. We seek training-free approaches of low-complexity, applicable to any foundation model and consider methods based on external vision-language model embeddings, like CLIP. This is denoted as Plug-and-Play Selective Prediction (PaPSP). We identify two key challenges: (1) instability of the visual-language representations, leading to high variance in image-text embeddings, and (2) poor calibration of similarity scores. To address these issues, we propose a memory augmented PaPSP (MA-PaPSP) model, which augments PaPSP with a retrieval dataset of image-text pairs. This is leveraged to reduce embedding variance by averaging retrieved nearest-neighbor pairs and is complemented by the use of contrastive normalization to improve score calibration. Through extensive experiments on multiple datasets, we show that MA-PaPSP outperforms PaPSP and other selective prediction baselines for selective captioning, image-text matching, and fine-grained classification. Code is publicly available at https://github.com/kingston-aditya/MA-PaPSP.	ICLR 2026	Code Link
Countering the Over-Reliance Trap: Mitigating Object Hallucination for LVLMs via a Self-Validation Framework	2026-01-30	Show Despite progress in Large Vision Language Models (LVLMs), object hallucination remains a critical issue in image captioning task, where models generate descriptions of non-existent objects, compromising their reliability. Previous work attributes this to LVLMs' over-reliance on language priors and attempts to mitigate it through logits calibration. However, they still lack a thorough analysis of the over-reliance. To gain a deeper understanding of over-reliance, we conduct a series of preliminary experiments, indicating that as the generation length increases, LVLMs' over-reliance on language priors leads to inflated probability of hallucinated object tokens, consequently exacerbating object hallucination. To circumvent this issue, we propose Language-Prior-Free Verification to enable LVLMs to faithfully verify the confidence of object existence. Based on this, we propose a novel training-free Self-Validation Framework to counter the over-reliance trap. It first validates objects' existence in sampled candidate captions and further mitigates object hallucination via caption selection or aggregation. Experiment results demonstrate that our framework mitigates object hallucination significantly in image captioning task (e.g., 65.6% improvement on CHAIRI metric with LLaVA-v1.5-7B), surpassing the previous SOTA methods. This result highlights a novel path towards mitigating hallucination by unlocking the inherent potential within LVLMs themselves.	Code ... Code is available at https://github.com/Liushiyu-0709/SelfVal	Code Link
Jailbreaks on Vision Language Model via Multimodal Reasoning	2026-01-29	Show Vision-language models (VLMs) have become central to tasks such as visual question answering, image captioning, and text-to-image generation. However, their outputs are highly sensitive to prompt variations, which can reveal vulnerabilities in safety alignment. In this work, we present a jailbreak framework that exploits post-training Chain-of-Thought (CoT) prompting to construct stealthy prompts capable of bypassing safety filters. To further increase attack success rates (ASR), we propose a ReAct-driven adaptive noising mechanism that iteratively perturbs input images based on model feedback. This approach leverages the ReAct paradigm to refine adversarial noise in regions most likely to activate safety defenses, thereby enhancing stealth and evasion. Experimental results demonstrate that the proposed dual-strategy significantly improves ASR while maintaining naturalness in both text and visual domains.		None
MultiModal Fine-tuning with Synthetic Captions	2026-01-29	Show In this paper, we address a fundamental gap between pre-training and fine-tuning of deep neural networks: while pre-training has shifted from unimodal to multimodal learning with enhanced visual understanding, fine-tuning predominantly remains unimodal, limiting the benefits of rich pre-trained representations. To bridge this gap, we propose a novel approach that transforms unimodal datasets into multimodal ones using Multimodal Large Language Models (MLLMs) to generate synthetic image captions for fine-tuning models with a multimodal objective. Our method employs carefully designed prompts incorporating class labels and domain context to produce high-quality captions tailored for classification tasks. Furthermore, we introduce a supervised contrastive loss function that explicitly encourages clustering of same-class representations during fine-tuning, along with a new inference technique that leverages class-averaged text embeddings from multiple synthetic captions per image. Extensive experiments across 13 image classification benchmarks demonstrate that our approach outperforms baseline methods, with particularly significant improvements in few-shot learning scenarios. Our work establishes a new paradigm for dataset enhancement that effectively bridges the gap between multimodal pre-training and fine-tuning. Our code is available at https://github.com/s-enmt/MMFT.		Code Link
bi-modal textual prompt learning for vision-language models in remote sensing	2026-01-28	Show Prompt learning (PL) has emerged as an effective strategy to adapt vision-language models (VLMs), such as CLIP, for downstream tasks under limited supervision. While PL has demonstrated strong generalization on natural image datasets, its transferability to remote sensing (RS) imagery remains underexplored. RS data present unique challenges, including multi-label scenes, high intra-class variability, and diverse spatial resolutions, that hinder the direct applicability of existing PL methods. In particular, current prompt-based approaches often struggle to identify dominant semantic cues and fail to generalize to novel classes in RS scenarios. To address these challenges, we propose BiMoRS, a lightweight bi-modal prompt learning framework tailored for RS tasks. BiMoRS employs a frozen image captioning model (e.g., BLIP-2) to extract textual semantic summaries from RS images. These captions are tokenized using a BERT tokenizer and fused with high-level visual features from the CLIP encoder. A lightweight cross-attention module then conditions a learnable query prompt on the fused textual-visual representation, yielding contextualized prompts without altering the CLIP backbone. We evaluate BiMoRS on four RS datasets across three domain generalization (DG) tasks and observe consistent performance gains, outperforming strong baselines by up to 2% on average. Codes are available at https://github.com/ipankhi/BiMoRS.	Accep... Accepted in ICASSP 2026	Code Link
GDCNet: Generative Discrepancy Comparison Network for Multimodal Sarcasm Detection	2026-01-28	Show Multimodal sarcasm detection (MSD) aims to identify sarcasm within image-text pairs by modeling semantic incongruities across modalities. Existing methods often exploit cross-modal embedding misalignment to detect inconsistency but struggle when visual and textual content are loosely related or semantically indirect. While recent approaches leverage large language models (LLMs) to generate sarcastic cues, the inherent diversity and subjectivity of these generations often introduce noise. To address these limitations, we propose the Generative Discrepancy Comparison Network (GDCNet). This framework captures cross-modal conflicts by utilizing descriptive, factually grounded image captions generated by Multimodal LLMs (MLLMs) as stable semantic anchors. Specifically, GDCNet computes semantic and sentiment discrepancies between the generated objective description and the original text, alongside measuring visual-textual fidelity. These discrepancy features are then fused with visual and textual representations via a gated module to adaptively balance modality contributions. Extensive experiments on MSD benchmarks demonstrate GDCNet's superior accuracy and robustness, establishing a new state-of-the-art on the MMSD2.0 benchmark.	Accep... Accepted to 2026 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2026)	None
Enhancing Descriptive Captions with Visual Attributes for Multimodal Perception	2026-01-27	Show Training Large Multimodality Models (LMMs) relies on descriptive image caption that connects image and language. Existing methods for generating such captions often rely on distilling the captions from pretrained LMMs, constructing them from publicly available internet images, or even generating them through human annotation. However, these strategies can fall short in terms of precision and granularity, particularly when dealing with complex visual reasoning tasks. In this paper, we propose to leverage off-the-shelf visual specialists, which were trained from annotated images initially not for image captioning, for enhancing the image caption. Our approach, named EDC, explores object low-level and fine-grained attributes (e.g., depth, emotion and fine-grained categories) and object relations (e.g., relative location and human-object-interaction (HOI)), and combine the attributes into the descriptive caption. By systematically integrating these rich attributes into the generated captions, EDC significantly improves the descriptive quality of the captions, providing a deeper and more nuanced understanding of the visual content. Experiments demonstrate that such visual specialists are able to improve the performance for visual understanding tasks as well as reasoning that benefits from more accurate visual understanding. The complete source code of EDC pipeline and datasets will be available at https://github.com/syp2ysy/DCE.	An op... An open-source Agent for generating detailed image captions	Code Link
ELIP: Efficient Discriminative Language-Image Pre-training with Fewer Vision Tokens	2026-01-26	Show Learning a versatile language-image model is computationally prohibitive under a limited computing budget. This paper delves into the \emph{efficient language-image pre-training}, an area that has received relatively little attention despite its importance in reducing computational cost and footprint. To that end, we propose a vision token pruning and merging method ELIP, to remove less influential tokens based on the supervision of language outputs. Our method is designed with several strengths, such as being computation-efficient, memory-efficient, and trainable-parameter-free, and is distinguished from previous vision-only token pruning approaches by its alignment with task objectives. We implement this method in a progressively pruning manner using several sequential blocks. To evaluate its generalization performance, we apply ELIP to three commonly used language-image pre-training models and utilize public image-caption pairs with 4M images for pre-training. Our experiments demonstrate that with the removal of ~30$%$ vision tokens across 12 ViT layers, ELIP maintains significantly comparable performance with baselines ($\sim$0.32 accuracy drop on average) over various downstream tasks including cross-modal retrieval, VQA, image captioning, \emph{etc}. In addition, the spared GPU resources by our ELIP allow us to scale up with larger batch sizes, thereby accelerating model pre-training and even sometimes enhancing downstream model performance.		None
Uni-RS: A Spatially Faithful Unified Understanding and Generation Model for Remote Sensing	2026-01-25	Show Unified remote sensing multimodal models exhibit a pronounced spatial reversal curse: Although they can accurately recognize and describe object locations in images, they often fail to faithfully execute the same spatial relations during text-to-image generation, where such relations constitute core semantic information in remote sensing. Motivated by this observation, we propose Uni-RS, the first unified multimodal model tailored for remote sensing, to explicitly address the spatial asymmetry between understanding and generation. Specifically, we first introduce explicit Spatial-Layout Planning to transform textual instructions into spatial layout plans, decoupling geometric planning from visual synthesis. We then impose Spatial-Aware Query Supervision to bias learnable queries toward spatial relations explicitly specified in the instruction. Finally, we develop Image-Caption Spatial Layout Variation to expose the model to systematic geometry-consistent spatial transformations. Extensive experiments across multiple benchmarks show that our approach substantially improves spatial faithfulness in text-to-image generation, while maintaining strong performance on multimodal understanding tasks like image captioning, visual grounding, and VQA tasks.		None
OpenVision 3: A Family of Unified Visual Encoder for Both Understanding and Generation	2026-01-21	Show This paper presents a family of advanced vision encoder, named OpenVision 3, that learns a single, unified visual representation that can serve both image understanding and image generation. Our core architecture is simple: we feed VAE-compressed image latents to a ViT encoder and train its output to support two complementary roles. First, the encoder output is passed to the ViT-VAE decoder to reconstruct the original image, encouraging the representation to capture generative structure. Second, the same representation is optimized with contrastive learning and image-captioning objectives, strengthening semantic features. By jointly optimizing reconstruction- and semantics-driven signals in a shared latent space, the encoder learns representations that synergize and generalize well across both regimes. We validate this unified design through extensive downstream evaluations with the encoder frozen. For multimodal understanding, we plug the encoder into the LLaVA-1.5 framework: it performs comparably with a standard CLIP vision encoder (e.g., 62.4 vs 62.2 on SeedBench, and 83.7 vs 82.9 on POPE). For generation, we test it under the RAE framework: ours substantially surpasses the standard CLIP-based encoder (e.g., gFID: 1.89 vs 2.54 on ImageNet). We hope this work can spur future research on unified modeling.		None
InstructTime++: Time Series Classification with Multimodal Language Modeling via Implicit Feature Enhancement	2026-01-21	Show Most existing time series classification methods adopt a discriminative paradigm that maps input sequences directly to one-hot encoded class labels. While effective, this paradigm struggles to incorporate contextual features and fails to capture semantic relationships among classes. To address these limitations, we propose InstructTime, a novel framework that reformulates time series classification as a multimodal generative task. Specifically, continuous numerical sequences, contextual textual features, and task instructions are treated as multimodal inputs, while class labels are generated as textual outputs by tuned language models. To bridge the modality gap, InstructTime introduces a time series discretization module that converts continuous sequences into discrete temporal tokens, together with an alignment projection layer and a generative self-supervised pre-training strategy to enhance cross-modal representation alignment. Building upon this framework, we further propose InstructTime++, which extends InstructTime by incorporating implicit feature modeling to compensate for the limited inductive bias of language models. InstructTime++ leverages specialized toolkits to mine informative implicit patterns from raw time series and contextual inputs, including statistical feature extraction and vision-language-based image captioning, and translates them into textual descriptions for seamless integration. Extensive experiments on multiple benchmark datasets demonstrate the superior performance of InstructTime++.		None
GAIA: A Global, Multi-modal, Multi-scale Vision-Language Dataset for Remote Sensing Image Analysis	2026-01-21	Show Existing Vision-Language Models (VLMs) are predominantly trained on web-scraped, noisy image-text data, exhibiting limited exposure to the specialized domain of RS. This deficiency results in poor performance on RS-specific tasks, as commonly used datasets often lack detailed, scientifically accurate textual descriptions and instead emphasize solely on attributes like date and location. To bridge this critical gap, we introduce GAIA, a novel dataset designed for multi-scale, multi-sensor, and multi-modal RS image analysis. GAIA comprises of 201,005 meticulously curated RS image-text pairs, representing a diverse range of RS modalities associated to different spatial resolutions. Unlike existing vision-language datasets in RS, GAIA specifically focuses on capturing a diverse range of RS applications, providing unique information about environmental changes, natural disasters, and various other dynamic phenomena. The dataset provides a spatially and temporally balanced distribution, spanning across the globe, covering the last 25 years with a balanced temporal distribution of observations. GAIA's construction involved a two-stage process: (1) targeted web-scraping of images and accompanying text from reputable RS-related sources, and (2) generation of five high-quality, scientifically grounded synthetic captions for each image using carefully crafted prompts that leverage the advanced vision-language capabilities of GPT-4o. Our extensive experiments, including fine-tuning of CLIP and BLIP2 models, demonstrate that GAIA significantly improves performance on RS image classification, cross-modal retrieval and image captioning tasks. We make our dataset, automated processing framework and fine-tuned model weights publicly available on our project's GitHub repository: https://github.com/Orion-AI-Lab/GAIA.	26 pages, 14 figures	Code Link
Hummus: A Dataset of Humorous Multimodal Metaphor Use	2026-01-20	Show Metaphor and humor share a lot of common ground, and metaphor is one of the most common humorous mechanisms. This study focuses on the humorous capacity of multimodal metaphors, which has not received due attention in the community. We take inspiration from the Incongruity Theory of humor, the Conceptual Metaphor Theory, and the annotation scheme behind the VU Amsterdam Metaphor Corpus, and developed a novel annotation scheme for humorous multimodal metaphor use in image-caption pairs. We create the Hummus Dataset of Humorous Multimodal Metaphor Use, providing expert annotation on 1k image-caption pairs sampled from the New Yorker Caption Contest corpus. Using the dataset, we test state-of-the-art multimodal large language models (MLLMs) on their ability to detect and understand humorous multimodal metaphor use. Our experiments show that current MLLMs still struggle with processing humorous multimodal metaphors, particularly with regard to integrating visual and textual information. We release our dataset and code at github.com/xiaoyuisrain/humorous-multimodal-metaphor-use.		Code Link
Dual-Stream Collaborative Transformer for Image Captioning	2026-01-19	Show Current region feature-based image captioning methods have progressed rapidly and achieved remarkable performance. However, they are still prone to generating irrelevant descriptions due to the lack of contextual information and the over-reliance on generated partial descriptions for predicting the remaining words. In this paper, we propose a Dual-Stream Collaborative Transformer (DSCT) to address this issue by introducing the segmentation feature. The proposed DSCT consolidates and then fuses the region and segmentation features to guide the generation of caption sentences. It contains multiple Pattern-Specific Mutual Attention Encoders (PSMAEs) and Dynamic Nomination Decoders (DNDs). The PSMAE effectively highlights and consolidates the private information of two representations by querying each other. The DND dynamically searches for the most relevant learning blocks to the input textual representations and exploits the homogeneous features between the consolidated region and segmentation features to generate more accurate and descriptive caption sentences. To the best of our knowledge, this is the first study to explore how to fuse different pattern-specific features in a dynamic way to bypass their semantic inconsistencies and spatial misalignment issues for image captioning. The experimental results from popular benchmark datasets demonstrate that our DSCT outperforms the state-of-the-art image captioning models in the literature.		None
RAPTOR-AI for Disaster OODA Loop: Hierarchical Multimodal RAG with Experience-Driven Agentic Decision-Making	2026-01-18	Show Effective humanitarian assistance and disaster relief (HADR) requires rapid situational understanding, reliable decision support, and the ability to generalize across diverse and previously unseen disaster contexts. This work introduces an agentic Retrieval-Augmented Generation (RAG) framework designed to support the three canonical phases of disaster response: initial rescue, mid-term recovery, and long-term reconstruction. To achieve robust multimodal grounding, we construct a hierarchical knowledge base that integrates textual disaster manuals, historical lessons (e.g., the 2011 Tohoku earthquake), and both aerial and ground-level imagery. Our system builds on the open-source multimodal implementation, which processes 46 tsunami-related PDFs (2,378 pages) using BLIP-based image captioning, ColVBERT embeddings, and long-context summarization to generate an efficient, structured multimodal retrieval tree optimized for disaster knowledge preservation. An agentic controller dynamically selects retrieval strategies (e.g., RAPTOR, ColBERT) through entropy-aware scene abstraction, enabling adaptive reasoning across heterogeneous inputs. Additionally, a lightweight LoRA-based post-training method injects experiential knowledge from past disasters, enhancing the models' capacity to support both expert and non-expert responders. Experiments on real disaster datasets demonstrate improved situational grounding, enhanced task decomposition accuracy, and superior usability for emergency operations. Incorporating recent advances in long-context RAG systems, agentic information retrieval, and contemporary emergency response AI, our system achieves substantial gains through adaptive retrieval-augmented generation with self-reasoning and multimodal chain-of-thought capabilities.	4 figures, 3 tables	None
LLM2CLIP: Powerful Language Model Unlocks Richer Visual Representation	2026-01-17	Show CLIP is a foundational multimodal model that aligns image and text features into a shared representation space via contrastive learning on large-scale image-text pairs. Its effectiveness primarily stems from the use of natural language as rich supervision. Motivated by the remarkable advancements in large language models (LLMs), this work explores how LLMs' superior text understanding and extensive open-world knowledge can enhance CLIP's capability, especially for processing longer and more complex image captions. We propose an efficient post-training strategy that integrates LLMs into pretrained CLIP. To address the challenge posed by the autoregressive nature of LLMs, we introduce a caption-to-caption contrastive fine-tuning framework, significantly enhancing the discriminative quality of LLM outputs. Extensive experiments demonstrate that our approach outperforms LoRA-based methods, achieving nearly fourfold faster training with superior performance. Furthermore, we validate substantial improvements over state-of-the-art models such as CLIP, EVA02, and SigLip2 across various zero-shot multimodal retrieval tasks, cross-lingual retrieval tasks, and multimodal language model pretraining.		None
Better Language Models Exhibit Higher Visual Alignment	2026-01-16	Show How well do text-only large language models (LLMs) align with the visual world? We present a systematic evaluation of this question by incorporating frozen representations of various language models into a discriminative vision-language framework and measuring zero-shot generalization to novel concepts. We find that decoder-based models exhibit stronger visual alignment than encoders, even when controlling for model and dataset size. Moreover, language modeling performance correlates with visual generalization, suggesting that advances in unimodal LLMs can simultaneously improve vision models. Leveraging these insights, we propose ShareLock, a lightweight method for fusing frozen vision and language backbones. ShareLock achieves robust performance across tasks while drastically reducing the need for paired data and compute. With just 563k image-caption pairs and under one GPU-hour of training, it reaches 51% accuracy on ImageNet. In cross-lingual settings, ShareLock dramatically outperforms CLIP, achieving 38.7% top-1 accuracy on Chinese image classification versus CLIP's 1.4%. Code is available.		None
DanQing: An Up-to-Date Large-Scale Chinese Vision-Language Pre-training Dataset	2026-01-15	Show Vision-Language Pre-training (VLP) models demonstrate strong performance across various downstream tasks by learning from large-scale image-text pairs through contrastive pretraining. The release of extensive English image-text datasets (e.g., COYO-700M and LAION-400M) has enabled widespread adoption of models such as CLIP and SigLIP in tasks including cross-modal retrieval and image captioning. However, the advancement of Chinese vision-language pretraining has substantially lagged behind, due to the scarcity of high-quality Chinese image-text data. To address this gap, we develop a comprehensive pipeline for constructing a high-quality Chinese cross-modal dataset. As a result, we propose DanQing, which contains 100 million image-text pairs collected from Common Crawl. Different from existing datasets, DanQing is curated through a more rigorous selection process, yielding superior data quality. Moreover, DanQing is primarily built from 2024-2025 web data, enabling models to better capture evolving semantic trends and thus offering greater practical utility. We compare DanQing with existing datasets by continual pre-training of the SigLIP2 model. Experimental results show that DanQing consistently achieves superior performance across a range of Chinese downstream tasks, including zero-shot classification, cross-modal retrieval, and LMM-based evaluations. To facilitate further research in Chinese vision-language pre-training, we will open-source the DanQing dataset under the Creative Common CC-BY 4.0 license.	19 pa... 19 pages, 11 figures, 7 tables	None
Multi-Modal LLM based Image Captioning in ICT: Bridging the Gap Between General and Industry Domain	2026-01-14	Show In the information and communications technology (ICT) industry, training a domain-specific large language model (LLM) or constructing a retrieval-augmented generation system requires a substantial amount of high-value domain knowledge. However, the knowledge is not only hidden in the textual modality but also in the image modality. Traditional methods can parse text from domain documents but dont have image captioning ability. Multi-modal LLM (MLLM) can understand images, but they do not have sufficient domain knowledge. To address the above issues, this paper proposes a multi-stage progressive training strategy to train a Domain-specific Image Captioning Model (DICModel) in ICT, and constructs a standard evaluation system to validate the performance of DICModel. Specifically, this work first synthesizes about 7K image-text pairs by combining the Mermaid tool and LLMs, which are used for the first-stage supervised-fine-tuning (SFT) of DICModel. Then, ICT-domain experts manually annotate about 2K image-text pairs for the second-stage SFT of DICModel. Finally, experts and LLMs jointly synthesize about 1.5K visual question answering data for the instruction-based SFT. Experimental results indicate that our DICModel with only 7B parameters performs better than other state-of-the-art models with 32B parameters. Compared to the SOTA models with 7B and 32B parameters, our DICModel increases the BLEU metric by approximately 56.8% and 20.8%, respectively. On the objective questions constructed by ICT domain experts, our DICModel outperforms Qwen2.5-VL 32B by 1% in terms of accuracy rate. In summary, this work can efficiently and accurately extract the logical text from images, which is expected to promote the development of multimodal models in the ICT domain.		None
Image-to-Brain Signal Generation for Visual Prosthesis with CLIP Guided Multimodal Diffusion Models	2026-01-14	Show Visual prostheses hold great promise for restoring vision in blind individuals. While researchers have successfully utilized M/EEG signals to evoke visual perceptions during the brain decoding stage of visual prostheses, the complementary process of converting images into M/EEG signals in the brain encoding stage remains largely unexplored, hindering the formation of a complete functional pipeline. In this work, we present a novel image-to-brain signal framework that generates M/EEG from images by leveraging the diffusion transformer architecture enhanced with cross-attention mechanisms. Specifically, we employ a diffusion transformer (DiT) architecture based on denoising diffusion implicit models (DDIM) to achieve brain signal generation. To realize the goal of image-to-brain signal conversion, we use cross-attention mechanisms to align brain signal embeddings with CLIP image embeddings. Moreover, we leverage large language models (LLMs) to generate image captions, and concatenate the resulting CLIP text embeddings with CLIP image embeddings to form unified embeddings for cross-attention alignment, enabling our model to capture core semantic information. Moreover, to capture core semantic information, we use large language models (LLMs) to generate descriptive and semantically accurate captions for images. Furthermore, we introduce a learnable spatio-temporal position encoding that combines brain region embeddings with temporal embeddings to capture both spatial and temporal characteristics of brain signals. We evaluate the framework on two multimodal benchmark datasets (THINGS-EEG2 and THINGS-MEG) and demonstrate that it generates biologically plausible brain signals.		None
GTR-CoT: Graph Traversal as Visual Chain of Thought for Molecular Structure Recognition	2026-01-13	Show Optical Chemical Structure Recognition (OCSR) is essential for converting molecular images into machine-readable formats. While recent vision-language models (VLMs) have shown promise, their image-captioning approach often struggles with complex molecular structures and inconsistent annotations. To address these issues, we introduce GTR-VL, featuring two key innovations: (1) the \textit{Graph Traversal as Visual Chain of Thought} mechanism that emulates human reasoning by incrementally parsing molecular graphs through sequential atom-bond predictions, and (2) the data-centric \textit{Faithfully Recognize What You've Seen} principle, which aligns abbreviated structures in images with their expanded annotations. For hand-drawn OCSR tasks, where datasets lack graph annotations and only provide final SMILES, we apply reinforcement learning using the GRPO method, introducing reward mechanisms like format reward, graph reward, and SMILES reward. This approach significantly enhances performance in hand-drawn recognition tasks through weak supervision. We developed GTR-1.3M, a large-scale instruction-tuning dataset with corrected annotations, and MolRec-Bench, the first benchmark for fine-grained evaluation of graph-parsing accuracy in OCSR. Our two-stage training scheme involves SFT training for printed images and the GRPO method for transferring capabilities to hand-drawn tasks. Experiments show that GTR-VL outperforms specialist models, chemistry-domain VLMs, and commercial VLMs on both printed and hand-drawn datasets.		None
UniF$^2$ace: A Unified Fine-grained Face Understanding and Generation Model	2026-01-13	Show Unified multimodal models (UMMs) have emerged as a powerful paradigm in fundamental cross-modality research, demonstrating significant potential in both image understanding and generation. However, existing research in the face domain primarily faces two challenges: $\textbf{(1)}$ $\textbf{fragmentation development}$, with existing methods failing to unify understanding and generation into a single one, hindering the way to artificial general intelligence. $\textbf{(2) lack of fine-grained facial attributes}$, which are crucial for high-fidelity applications. To handle those issues, we propose $\textbf{UniF$^2$ace}$, $\textit{the first UMM specifically tailored for fine-grained face understanding and generation}$. $\textbf{First}$, we introduce a novel theoretical framework with a Dual Discrete Diffusion (D3Diff) loss, unifying masked generative models with discrete score matching diffusion and leading to a more precise approximation of the negative log-likelihood. Moreover, this D3Diff significantly enhances the model's ability to synthesize high-fidelity facial details aligned with text input. $\textbf{Second}$, we propose a multi-level grouped Mixture-of-Experts architecture, adaptively incorporating the semantic and identity facial embeddings to complement the attribute forgotten phenomenon in representation evolvement. $\textbf{Finally}$, to this end, we construct UniF$^2$aceD-1M, a large-scale dataset comprising 130K fine-grained image-caption pairs and 1M visual question-answering pairs, spanning a much wider range of facial attributes than existing datasets. Extensive experiments demonstrate that UniF$^2$ace outperforms existing models with a similar scale in both understanding and generation tasks, with 7.1% higher Desc-GPT and 6.6% higher VQA-score, respectively.		None
OSCAR: Open-Set CAD Retrieval from a Language Prompt and a Single Image	2026-01-12	Show 6D object pose estimation plays a crucial role in scene understanding for applications such as robotics and augmented reality. To support the needs of ever-changing object sets in such context, modern zero-shot object pose estimators were developed to not require object-specific training but only rely on CAD models. Such models are hard to obtain once deployed, and a continuously changing and growing set of objects makes it harder to reliably identify the instance model of interest. To address this challenge, we introduce an Open-Set CAD Retrieval from a Language Prompt and a Single Image (OSCAR), a novel training-free method that retrieves a matching object model from an unlabeled 3D object database. During onboarding, OSCAR generates multi-view renderings of database models and annotates them with descriptive captions using an image captioning model. At inference, GroundedSAM detects the queried object in the input image, and multi-modal embeddings are computed for both the Region-of-Interest and the database captions. OSCAR employs a two-stage retrieval: text-based filtering using CLIP identifies candidate models, followed by image-based refinement using DINOv2 to select the most visually similar object. In our experiments we demonstrate that OSCAR outperforms all state-of-the-art methods on the cross-domain 3D model retrieval benchmark MI3DOR. Furthermore, we demonstrate OSCAR's direct applicability in automating object model sourcing for 6D object pose estimation. We propose using the most similar object model for pose estimation if the exact instance is not available and show that OSCAR achieves an average precision of 90.48% during object retrieval on the YCB-V object dataset. Moreover, we demonstrate that the most similar object model can be utilized for pose estimation using Megapose achieving better results than a reconstruction-based approach.		None
Context-Aware Decoding for Faithful Vision-Language Generation	2026-01-09	Show Hallucinations, generating responses inconsistent with the visual input, remain a critical limitation of large vision-language models (LVLMs), especially in open-ended tasks such as image captioning and visual reasoning. In this work, we probe the layer-wise generation dynamics that drive hallucinations and propose a training-free mitigation strategy. Employing the Logit Lens, we examine how LVLMs construct next-token distributions across decoder layers, uncovering a pronounced commitment-depth gap: truthful tokens accumulate probability mass on their final candidates earlier than hallucinatory ones. Drawing on this discovery, we introduce Context Embedding Injection (CEI), a lightweight method that harnesses the hidden state of the last input token-the context embedding-as a grounding signal to maintain visual fidelity throughout decoding and curb hallucinations. Evaluated on the CHAIR, AMBER, and MMHal-Bench benchmarks (with a maximum token length of 512), CEI outperforms state-of-the-art baselines across three LVLMs, with its dynamic variant yielding the lowest overall hallucination rates. By integrating novel mechanistic insights with a scalable intervention, this work advances the mitigation of hallucinations in LVLMs.		None
TDHook: A Lightweight Framework for Interpretability	2026-01-09	Show Interpretability of Deep Neural Networks (DNNs) is a growing field driven by the study of vision and language models. Yet, some use cases, like image captioning, or domains like Deep Reinforcement Learning (DRL), require complex modelling, with multiple inputs and outputs or use composable and separated networks. As a consequence, they rarely fit natively into the API of popular interpretability frameworks. We thus present TDHook, an open-source, lightweight, generic interpretability framework based on $\texttt{tensordict}$ and applicable to any $\texttt{torch}$ model. It focuses on handling complex composed models which can be trained for Computer Vision, Natural Language Processing, Reinforcement Learning or any other domain. This library features ready-to-use methods for attribution, probing and a flexible get-set API for interventions, and is aiming to bridge the gap between these method classes to make modern interpretability pipelines more accessible. TDHook is designed with minimal dependencies, requiring roughly half as much disk space as $\texttt{transformer_lens}$, and, in our controlled benchmark, achieves up to a $\times$2 speed-up over $\texttt{captum}$ when running integrated gradients for multi-target pipelines on both CPU and GPU. In addition, to value our work, we showcase concrete use cases of our library with composed interpretability pipelines in Computer Vision (CV) and Natural Language Processing (NLP), as well as with complex models in DRL.		None
Unified Text-Image Generation with Weakness-Targeted Post-Training	2026-01-07	Show Unified multimodal generation architectures that jointly produce text and images have recently emerged as a promising direction for text-to-image (T2I) synthesis. However, many existing systems rely on explicit modality switching, generating reasoning text before switching manually to image generation. This separate, sequential inference process limits cross-modal coupling and prohibits automatic multimodal generation. This work explores post-training to achieve fully unified text-image generation, where models autonomously transition from textual reasoning to visual synthesis within a single inference process. We examine the impact of joint text-image generation on T2I performance and the relative importance of each modality during post-training. We additionally explore different post-training data strategies, showing that a targeted dataset addressing specific limitations achieves superior results compared to broad image-caption corpora or benchmark-aligned data. Using offline, reward-weighted post-training with fully self-generated synthetic data, our approach enables improvements in multimodal image generation across four diverse T2I benchmarks, demonstrating the effectiveness of reward-weighting both modalities and strategically designed post-training data.		None
OmniNav: A Unified Framework for Prospective Exploration and Visual-Language Navigation	2026-01-07	Show Embodied navigation presents a core challenge for intelligent robots, requiring the comprehension of visual environments, natural language instructions, and autonomous exploration. Existing models often fall short in offering a unified solution across diverse navigation paradigms, resulting in low success rates and limited generalization. We introduce OmniNav, a unified framework addressing instruct-goal, object-goal, point-goal navigation, and frontier-based exploration within a single architecture. Our approach features a lightweight, low-latency policy that accurately predicts continuous-space waypoints (coordinates and orientations). This policy surpasses action-chunk methods in precision and supports real-world deployment at control frequencies up to 5 Hz. Architecturally, OmniNav employs a fast-slow system design: a fast module generates waypoints using short-horizon visual context and subtasks, while a slow module performs deliberative planning with long-horizon observations and candidate frontiers to select subsequent subgoals and subtasks. This collaboration enhances path efficiency and maintains trajectory coherence, particularly in exploration and memory-intensive scenarios. Crucially, we identify that the primary bottleneck isn't merely navigation policy learning, but a robust understanding of general instructions and objects. To boost generalization, OmniNav integrates large-scale, general-purpose training datasets, including those for image captioning and visual recognition, into a joint multi-task regimen. This significantly improves success rates and robustness. Extensive experiments confirm OmniNav's state-of-the-art performance across various navigation benchmarks, with real-world deployment further validating its efficacy. OmniNav provides practical insights for embodied navigation, charting a scalable path towards versatile, highly generalizable robotic intelligence.		None
RxnCaption: Reformulating Reaction Diagram Parsing as Visual Prompt Guided Captioning	2026-01-06	Show Large-scale chemical reaction datasets are crucial for AI research in chemistry. However, existing chemical reaction data often exist as images within papers, making them not machine-readable and unusable for training machine learning models. In response to this challenge, we propose the RxnCaption framework for the task of chemical Reaction Diagram Parsing (RxnDP). Our framework reformulates the traditional coordinate prediction driven parsing process into an image captioning problem, which Large Vision Language Models (LVLMs) handle naturally. We introduce a strategy termed BBox and Index as Visual Prompt (BIVP), which uses our state-of-the-art molecular detector, MolYOLO, to pre-draw molecular bounding boxes and indices directly onto the input image. This turns the downstream parsing into a natural-language description problem. Extensive experiments show that the BIVP strategy significantly improves structural extraction quality while simplifying model design. We further construct the RxnCaption-15k dataset, an order of magnitude larger than prior real-world literature benchmarks, with a balanced test subset across four layout archetypes. Experiments demonstrate that RxnCaption-VL achieves state-of-the-art performance on multiple metrics. We believe our method, dataset, and models will advance structured information extraction from chemical literature and catalyze broader AI applications in chemistry. We will release data, models, and code on GitHub.		None
Entity-Guided Multi-Task Learning for Infrared and Visible Image Fusion	2026-01-05	Show Existing text-driven infrared and visible image fusion approaches often rely on textual information at the sentence level, which can lead to semantic noise from redundant text and fail to fully exploit the deeper semantic value of textual information. To address these issues, we propose a novel fusion approach named Entity-Guided Multi-Task learning for infrared and visible image fusion (EGMT). Our approach includes three key innovative components: (i) A principled method is proposed to extract entity-level textual information from image captions generated by large vision-language models, eliminating semantic noise from raw text while preserving critical semantic information; (ii) A parallel multi-task learning architecture is constructed, which integrates image fusion with a multi-label classification task. By using entities as pseudo-labels, the multi-label classification task provides semantic supervision, enabling the model to achieve a deeper understanding of image content and significantly improving the quality and semantic density of the fused image; (iii) An entity-guided cross-modal interactive module is also developed to facilitate the fine-grained interaction between visual and entity-level textual features, which enhances feature representation by capturing cross-modal dependencies at both inter-visual and visual-entity levels. To promote the wide application of the entity-guided image fusion framework, we release the entity-annotated version of four public datasets (i.e., TNO, RoadScene, M3FD, and MSRS). Extensive experiments demonstrate that EGMT achieves superior performance in preserving salient targets, texture details, and semantic consistency, compared to the state-of-the-art methods. The code and dataset will be publicly available at https://github.com/wyshao-01/EGMT.	Accep... Accepted by IEEE Transactions on Multimedia	Code Link
AdaptInfer: Adaptive Token Pruning for Vision-Language Model Inference with Dynamical Text Guidance	2026-01-05	Show Vision-language models (VLMs) have achieved impressive performance on multimodal reasoning tasks such as visual question answering, image captioning and so on, but their inference cost remains a significant challenge due to the large number of vision tokens processed during the prefill stage. Existing pruning methods often rely on directly using the attention patterns or static text prompt guidance, failing to exploit the dynamic internal signals generated during inference. To address these issues, we propose AdaptInfer, a plug-and-play framework for adaptive vision token pruning in VLMs. First, we introduce a fine-grained, dynamic text-guided pruning mechanism that reuses layer-wise text-to-text attention maps to construct soft priors over text-token importance, allowing more informed scoring of vision tokens at each stage. Second, we perform an offline analysis of cross-modal attention shifts and identify consistent inflection locations in inference, which inspire us to propose a more principled and efficient pruning schedule. Our method is lightweight and plug-and-play, also generalizable across multi-modal tasks. Experimental results have verified the effectiveness of the proposed method. For example, it reduces CUDA latency by 61.3% while maintaining an average accuracy of 93.1% on vanilla LLaVA-1.5-7B. Under the same token budget, AdaptInfer surpasses SOTA in accuracy.		None
Towards Vision-Language Geo-Foundation Model: A Survey	2026-01-04	Show Vision-Language Foundation Models (VLFMs) have made remarkable progress on various multimodal tasks, such as image captioning, image-text retrieval, visual question answering, and visual grounding. However, most methods rely on training with general image datasets, and the lack of geospatial data leads to poor performance on earth observation. Numerous geospatial image-text pair datasets and VLFMs fine-tuned on them have been proposed recently. These new approaches aim to leverage large-scale, multimodal geospatial data to build versatile intelligent models with diverse geo-perceptive capabilities, which we refer to as Vision-Language Geo-Foundation Models (VLGFMs). This paper thoroughly reviews VLGFMs, summarizing and analyzing recent developments in the field. In particular, we introduce the background and motivation behind the rise of VLGFMs, highlighting their unique research significance. Then, we systematically summarize the core technologies employed in VLGFMs, including data construction, model architectures, and applications of various multimodal geospatial tasks. Finally, we conclude with insights, issues, and discussions regarding future research directions. To the best of our knowledge, this is the first comprehensive literature review of VLGFMs. We keep tracing related works at https://github.com/zytx121/Awesome-VLGFM.	18 pages, 4 figures	Code Link
SPoRC-VIST: A Benchmark for Evaluating Generative Natural Narrative in Vision-Language Models	2026-01-03	Show Vision-Language Models (VLMs) have achieved remarkable success in descriptive tasks such as image captioning and visual question answering (VQA). However, their ability to generate engaging, long-form narratives -- specifically multi-speaker podcast dialogues -- remains under-explored and difficult to evaluate. Standard metrics like BLEU and ROUGE fail to capture the nuances of conversational naturalness, personality, and narrative flow, often rewarding safe, repetitive outputs over engaging storytelling. In this work, we present a novel pipeline for end-to-end visual podcast generation, and fine-tune a Qwen3-VL-32B model on a curated dataset of 4,000 image-dialogue pairs. Crucially, we use a synthetic-to-real training strategy: we train on high-quality podcast dialogues from the Structured Podcast Research Corpus (SPoRC) paired with synthetically generated imagery, and evaluate on real-world photo sequences from the Visual Storytelling Dataset (VIST). This rigorous setup tests the model's ability to generalize from synthetic training data to real-world visual domains. We propose a comprehensive evaluation framework that moves beyond textual overlap, and use AI-as-a-judge (Gemini 3 Pro, Claude Opus 4.5, GPT 5.2) and novel style metrics (average turn length, speaker switch rate) to assess quality. Our experiments demonstrate that our fine-tuned 32B model significantly outperforms a 235B base model in conversational naturalness ($>$80% win rate) and narrative depth (+50% turn length), while maintaining identical visual grounding capabilities (CLIPScore: 20.39).	14 pa... 14 pages, 3 figures. Accepted to WVAQ 2026, WACV 2026	None
VisualQuest: A Benchmark for Abstract Visual Reasoning in MLLMs	2026-01-02	Show We introduce VisualQuest, a novel dataset designed to rigorously evaluate multimodal large language models (MLLMs) on abstract visual reasoning tasks that require the integration of symbolic, cultural, and linguistic knowledge. Unlike existing benchmarks that focus on direct image captioning or classification of realistic images, VisualQuest comprises 3,551 non-photographic, stylized images spanning four categories: Public Figures, Popular Culture, Linguistic Expressions, and Literary Works. Each image is paired with targeted questions to probe complex reasoning. We benchmark ten state-of-the-art MLLMs and find that only Gemini-2.5-flash and GPT-4o achieve strong overall performance, while 3.7 percent of the images remain unrecognized by any model, underscoring persistent challenges in multimodal understanding. Fine-grained analysis shows that Gemini excels at recognizing stylized public figures, whereas GPT-4o leads in linguistic reasoning tasks such as visual puns and emoji combinations. VisualQuest provides a comprehensive and challenging resource for advancing research in abstract visual reasoning and highlights key areas for future model improvement. The dataset is available at https://github.com/xkt88/VISUALQUEST.		Code Link
CPJ: Explainable Agricultural Pest Diagnosis via Caption-Prompt-Judge with LLM-Judged Refinement	2025-12-31	Show Accurate and interpretable crop disease diagnosis is essential for agricultural decision-making, yet existing methods often rely on costly supervised fine-tuning and perform poorly under domain shifts. We propose Caption--Prompt--Judge (CPJ), a training-free few-shot framework that enhances Agri-Pest VQA through structured, interpretable image captions. CPJ employs large vision-language models to generate multi-angle captions, refined iteratively via an LLM-as-Judge module, which then inform a dual-answer VQA process for both recognition and management responses. Evaluated on CDDMBench, CPJ significantly improves performance: using GPT-5-mini captions, GPT-5-Nano achieves \textbf{+22.7} pp in disease classification and \textbf{+19.5} points in QA score over no-caption baselines. The framework provides transparent, evidence-based reasoning, advancing robust and explainable agricultural diagnosis without fine-tuning. Our code and data are publicly available at: https://github.com/CPJ-Agricultural/CPJ-Agricultural-Diagnosis.	This ... This paper is 6 pages in length and contains 2 figures. Tao Fang (Corresponding Author), Lina Lu (Co-corresponding Author)	Code Link
ReVision: A Dataset and Baseline VLM for Privacy-Preserving Task-Oriented Visual Instruction Rewriting	2025-12-31	Show Efficient and privacy-preserving multimodal interaction is essential as AR, VR, and modern smartphones with powerful cameras become primary interfaces for human-computer communication. Existing powerful large vision-language models (VLMs) enabling multimodal interaction often rely on cloud-based processing, raising significant concerns about (1) visual privacy by transmitting sensitive vision data to servers, and (2) their limited real-time, on-device usability. This paper explores Visual Instruction Rewriting, a novel approach that transforms multimodal instructions into text-only commands, allowing seamless integration of lightweight on-device instruction rewriter VLMs (250M parameters) with existing conversational AI systems, enhancing vision data privacy. To achieve this, we present a dataset of over 39,000 examples across 14 domains and develop a compact VLM, pretrained on image captioning datasets and fine-tuned for instruction rewriting. Experimental results, evaluated through NLG metrics such as BLEU, METEOR, and ROUGE, along with semantic parsing analysis, demonstrate that even a quantized version of the model (<500MB storage footprint) can achieve effective instruction rewriting, thus enabling privacy-focused, multimodal AI applications.	Accep... Accepted and to appear in IJCNLP-AACL 2025	None
FUSE-RSVLM: Feature Fusion Vision-Language Model for Remote Sensing	2025-12-30	Show Large vision-language models (VLMs) exhibit strong performance across various tasks. However, these VLMs encounter significant challenges when applied to the remote sensing domain due to the inherent differences between remote sensing images and natural images. Existing remote sensing VLMs often fail to extract fine-grained visual features and suffer from visual forgetting during deep language processing. To address this, we introduce MF-RSVLM, a Multi-Feature Fusion Remote Sensing Vision--Language Model that effectively extracts and fuses visual features for RS understanding. MF-RSVLM learns multi-scale visual representations and combines global context with local details, improving the capture of small and complex structures in RS scenes. A recurrent visual feature injection scheme ensures the language model remains grounded in visual evidence and reduces visual forgetting during generation. Extensive experiments on diverse RS benchmarks show that MF-RSVLM achieves state-of-the-art or highly competitive performance across remote sensing classification, image captioning, and VQA tasks. Our code is publicly available at https://github.com/Yunkaidang/RSVLM.		Code Link
Scaling Remote Sensing Foundation Models: Data Domain Tradeoffs at the Peta-Scale	2025-12-29	Show We explore the scaling behaviors of artificial intelligence to establish practical techniques for training foundation models on high-resolution electro-optical (EO) datasets that exceed the current state-of-the-art scale by orders of magnitude. Modern multimodal machine learning (ML) applications, such as generative artificial intelligence (GenAI) systems for image captioning, search, and reasoning, depend on robust, domain-specialized encoders for non-text modalities. In natural-image domains where internet-scale data is plentiful, well-established scaling laws help optimize the joint scaling of model capacity, training compute, and dataset size. Unfortunately, these relationships are much less well-understood in high-value domains like remote sensing (RS). Using over a quadrillion pixels of commercial satellite EO data and the MITRE Federal AI Sandbox, we train progressively larger vision transformer (ViT) backbones, report success and failure modes observed at petascale, and analyze implications for bridging domain gaps across additional RS modalities. We observe that even at this scale, performance is consistent with a data limited regime rather than a model parameter-limited one. These practical insights are intended to inform data-collection strategies, compute budgets, and optimization schedules that advance the future development of frontier-scale RS foundation models.		None
Multimodal Interpretation of Remote Sensing Images: Dynamic Resolution Input Strategy and Multi-scale Vision-Language Alignment Mechanism	2025-12-29	Show Multimodal fusion of remote sensing images serves as a core technology for overcoming the limitations of single-source data and improving the accuracy of surface information extraction, which exhibits significant application value in fields such as environmental monitoring and urban planning. To address the deficiencies of existing methods, including the failure of fixed resolutions to balance efficiency and detail, as well as the lack of semantic hierarchy in single-scale alignment, this study proposes a Vision-language Model (VLM) framework integrated with two key innovations: the Dynamic Resolution Input Strategy (DRIS) and the Multi-scale Vision-language Alignment Mechanism (MS-VLAM).Specifically, the DRIS adopts a coarse-to-fine approach to adaptively allocate computational resources according to the complexity of image content, thereby preserving key fine-grained features while reducing redundant computational overhead. The MS-VLAM constructs a three-tier alignment mechanism covering object, local-region and global levels, which systematically captures cross-modal semantic consistency and alleviates issues of semantic misalignment and granularity imbalance.Experimental results on the RS-GPT4V dataset demonstrate that the proposed framework significantly improves the accuracy of semantic understanding and computational efficiency in tasks including image captioning and cross-modal retrieval. Compared with conventional methods, it achieves superior performance in evaluation metrics such as BLEU-4 and CIDEr for image captioning, as well as R@10 for cross-modal retrieval. This technical framework provides a novel approach for constructing efficient and robust multimodal remote sensing systems, laying a theoretical foundation and offering technical guidance for the engineering application of intelligent remote sensing interpretation.		None
View Selection for 3D Captioning via Diffusion Ranking	2025-12-27	Show Scalable annotation approaches are crucial for constructing extensive 3D-text datasets, facilitating a broader range of applications. However, existing methods sometimes lead to the generation of hallucinated captions, compromising caption quality. This paper explores the issue of hallucination in 3D object captioning, with a focus on Cap3D method, which renders 3D objects into 2D views for captioning using pre-trained models. We pinpoint a major challenge: certain rendered views of 3D objects are atypical, deviating from the training data of standard image captioning models and causing hallucinations. To tackle this, we present DiffuRank, a method that leverages a pre-trained text-to-3D model to assess the alignment between 3D objects and their 2D rendered views, where the view with high alignment closely represent the object's characteristics. By ranking all rendered views and feeding the top-ranked ones into GPT4-Vision, we enhance the accuracy and detail of captions, enabling the correction of 200k captions in the Cap3D dataset and extending it to 1 million captions across Objaverse and Objaverse-XL datasets. Additionally, we showcase the adaptability of DiffuRank by applying it to pre-trained text-to-image models for a Visual Question Answering task, where it outperforms the CLIP model.	Datas... Dataset link: https://huggingface.co/datasets/tiange/Cap3D	None
LLM-Free Image Captioning Evaluation in Reference-Flexible Settings	2025-12-25	Show We focus on the automatic evaluation of image captions in both reference-based and reference-free settings. Existing metrics based on large language models (LLMs) favor their own generations; therefore, the neutrality is in question. Most LLM-free metrics do not suffer from such an issue, whereas they do not always demonstrate high performance. To address these issues, we propose Pearl, an LLM-free supervised metric for image captioning, which is applicable to both reference-based and reference-free settings. We introduce a novel mechanism that learns the representations of image--caption and caption--caption similarities. Furthermore, we construct a human-annotated dataset for image captioning metrics, that comprises approximately 333k human judgments collected from 2,360 annotators across over 75k images. Pearl outperformed other existing LLM-free metrics on the Composite, Flickr8K-Expert, Flickr8K-CF, Nebula, and FOIL datasets in both reference-based and reference-free settings. Our project page is available at https://pearl.kinsta.page/.	Accep... Accepted for presentation at AAAI2026	None
VisRes Bench: On Evaluating the Visual Reasoning Capabilities of VLMs	2025-12-24	Show Vision-Language Models (VLMs) have achieved remarkable progress across tasks such as visual question answering and image captioning. Yet, the extent to which these models perform visual reasoning as opposed to relying on linguistic priors remains unclear. To address this, we introduce VisRes Bench, a benchmark designed to study visual reasoning in naturalistic settings without contextual language supervision. Analyzing model behavior across three levels of complexity, we uncover clear limitations in perceptual and relational visual reasoning capacities. VisRes isolates distinct reasoning abilities across its levels. Level 1 probes perceptual completion and global image matching under perturbations such as blur, texture changes, occlusion, and rotation; Level 2 tests rule-based inference over a single attribute (e.g., color, count, orientation); and Level 3 targets compositional reasoning that requires integrating multiple visual attributes. Across more than 19,000 controlled task images, we find that state-of-the-art VLMs perform near random under subtle perceptual perturbations, revealing limited abstraction beyond pattern recognition. We conclude by discussing how VisRes provides a unified framework for advancing abstract visual reasoning in multimodal research.		None
Proprioception Enhances Vision Language Model in Generating Captions and Subtask Segmentations for Robot Task	2025-12-24	Show From the perspective of future developments in robotics, it is crucial to verify whether foundation models trained exclusively on offline data, such as images and language, can understand the robot motion. In particular, since Vision Language Models (VLMs) do not include low-level motion information from robots in their training datasets, video understanding including trajectory information remains a significant challenge. In this study, we assess two capabilities of VLMs through a video captioning task with low-level robot motion information: (1) automatic captioning of robot tasks and (2) segmentation of a series of tasks. Both capabilities are expected to enhance the efficiency of robot imitation learning by linking language and motion and serve as a measure of the foundation model's performance. The proposed method generates multiple "scene" captions using image captions and trajectory data from robot tasks. The full task caption is then generated by summarizing these individual captions. Additionally, the method performs subtask segmentation by comparing the similarity between text embeddings of image captions. In both captioning tasks, the proposed method aims to improve performance by providing the robot's motion data - joint and end-effector states - as input to the VLM. Simulator experiments were conducted to validate the effectiveness of the proposed method.		None
Beyond Vision: Contextually Enriched Image Captioning with Multi-Modal Retrieva	2025-12-23	Show Real-world image captions often lack contextual depth, omitting crucial details such as event background, temporal cues, outcomes, and named entities that are not visually discernible. This gap limits the effectiveness of image understanding in domains like journalism, education, and digital archives, where richer, more informative descriptions are essential. To address this, we propose a multimodal pipeline that augments visual input with external textual knowledge. Our system retrieves semantically similar images using BEIT-3 (Flickr30k-384 and COCO-384) and SigLIP So-384, reranks them using ORB and SIFT for geometric alignment, and extracts contextual information from related articles via semantic search. A fine-tuned Qwen3 model with QLoRA then integrates this context with base captions generated by Instruct BLIP (Vicuna-7B) to produce event-enriched, context-aware descriptions. Evaluated on the OpenEvents v1 dataset, our approach generates significantly more informative captions compared to traditional methods, showing strong potential for real-world applications requiring deeper visual-textual understanding	7 pag... 7 pages, 5 figures. System description for the EVENTA Grand Challenge (Track 1) at ACM MM'25	None
Brain-language fusion enables interactive neural readout and in-silico experimentation	2025-12-22	Show Large language models (LLMs) have revolutionized human-machine interaction, and have been extended by embedding diverse modalities such as images into a shared language space. Yet, neural decoding has remained constrained by static, non-interactive methods. We introduce CorText, a framework that integrates neural activity directly into the latent space of an LLM, enabling open-ended, natural language interaction with brain data. Trained on fMRI data recorded during viewing of natural scenes, CorText generates accurate image captions and can answer more detailed questions better than controls, while having access to neural data only. We showcase that CorText achieves zero-shot generalization beyond semantic categories seen during training. In-silico microstimulation experiments, which enable counterfactual prompts on brain activity, reveal a consistent, and graded mapping between brain-state and language output. These advances mark a shift from passive decoding toward generative, flexible interfaces between brain activity and language.	v2	None
A Benchmark for Ultra-High-Resolution Remote Sensing MLLMs	2025-12-19	Show Multimodal large language models (MLLMs) demonstrate strong perception and reasoning performance on existing remote sensing (RS) benchmarks. However, most prior benchmarks rely on low-resolution imagery, and some high-resolution benchmarks suffer from flawed reasoning-task designs. We show that text-only LLMs can perform competitively with multimodal vision-language models on RS reasoning tasks without access to images, revealing a critical mismatch between current benchmarks and the intended evaluation of visual understanding. To enable faithful assessment, we introduce RSHR-Bench, a super-high-resolution benchmark for RS visual understanding and reasoning. RSHR-Bench contains 5,329 full-scene images with a long side of at least 4,000 pixels, with up to about 3 x 10^8 pixels per image, sourced from widely used RS corpora and UAV collections. We design four task families: multiple-choice VQA, open-ended VQA, image captioning, and single-image evaluation. These tasks cover nine perception categories and four reasoning types, supporting multi-turn and multi-image dialog. To reduce reliance on language priors, we apply adversarial filtering with strong LLMs followed by rigorous human verification. Overall, we construct 3,864 VQA tasks, 3,913 image captioning tasks, and 500 fully human-written or verified single-image evaluation VQA pairs. Evaluations across open-source, closed-source, and RS-specific VLMs reveal persistent performance gaps in super-high-resolution scenarios. Code: https://github.com/Yunkaidang/RSHR		Code Link
Auto-Vocabulary 3D Object Detection	2025-12-18	Show Open-vocabulary 3D object detection methods are able to localize 3D boxes of classes unseen during training. Despite the name, existing methods rely on user-specified classes both at training and inference. We propose to study Auto-Vocabulary 3D Object Detection (AV3DOD), where the classes are automatically generated for the detected objects without any user input. To this end, we introduce Semantic Score (SS) to evaluate the quality of the generated class names. We then develop a novel framework, AV3DOD, which leverages 2D vision-language models (VLMs) to generate rich semantic candidates through image captioning, pseudo 3D box generation, and feature-space semantics expansion. AV3DOD achieves the state-of-the-art (SOTA) performance on both localization (mAP) and semantic quality (SS) on the ScanNetV2 and SUNRGB-D datasets. Notably, it surpasses the SOTA, CoDA, by 3.48 overall mAP and attains a 24.5% relative improvement in SS on ScanNetV2.	technical report	None
An Efficient and Effective Encoder Model for Vision and Language Tasks in the Remote Sensing Domain	2025-12-17	Show The remote sensing community has recently seen the emergence of methods based on Large Vision and Language Models (LVLMs) that can address multiple tasks at the intersection of computer vision and natural language processing. To fully exploit the potential of such models, a significant focus has been given to the collection of large amounts of training data that cover multiple remote sensing-specific tasks, such as image captioning or visual question answering. However, the cost of using and training LVLMs is high, due to the large number of parameters. While multiple parameter-efficient adaptation techniques have been explored, the computational costs of training and inference with these models can remain prohibitive for most institutions. In this work, we explore the use of encoder-only architectures and propose a model that can effectively address multi-task learning while remaining compact in terms of the number of parameters. In particular, our model tackles combinations of tasks that are not typically explored in a unified model: the generation of text from remote sensing images and cross-modal retrieval. The results of our GeoMELT model - named from Multi-task Efficient Learning Transformer - in established benchmarks confirm the efficacy and efficiency of the proposed approach.		None
ComMark: Covert and Robust Black-Box Model Watermarking with Compressed Samples	2025-12-16	Show The rapid advancement of deep learning has turned models into highly valuable assets due to their reliance on massive data and costly training processes. However, these models are increasingly vulnerable to leakage and theft, highlighting the critical need for robust intellectual property protection. Model watermarking has emerged as an effective solution, with black-box watermarking gaining significant attention for its practicality and flexibility. Nonetheless, existing black-box methods often fail to better balance covertness (hiding the watermark to prevent detection and forgery) and robustness (ensuring the watermark resists removal)-two essential properties for real-world copyright verification. In this paper, we propose ComMark, a novel black-box model watermarking framework that leverages frequency-domain transformations to generate compressed, covert, and attack-resistant watermark samples by filtering out high-frequency information. To further enhance watermark robustness, our method incorporates simulated attack scenarios and a similarity loss during training. Comprehensive evaluations across diverse datasets and architectures demonstrate that ComMark achieves state-of-the-art performance in both covertness and robustness. Furthermore, we extend its applicability beyond image recognition to tasks including speech recognition, sentiment analysis, image generation, image captioning, and video recognition, underscoring its versatility and broad applicability.		None
DISCODE: Distribution-Aware Score Decoder for Robust Automatic Evaluation of Image Captioning	2025-12-16	Show Large vision-language models (LVLMs) have shown impressive performance across a broad range of multimodal tasks. However, robust image caption evaluation using LVLMs remains challenging, particularly under domain-shift scenarios. To address this issue, we introduce the Distribution-Aware Score Decoder (DISCODE), a novel finetuning-free method that generates robust evaluation scores better aligned with human judgments across diverse domains. The core idea behind DISCODE lies in its test-time adaptive evaluation approach, which introduces the Adaptive Test-Time (ATT) loss, leveraging a Gaussian prior distribution to improve robustness in evaluation score estimation. This loss is efficiently minimized at test time using an analytical solution that we derive. Furthermore, we introduce the Multi-domain Caption Evaluation (MCEval) benchmark, a new image captioning evaluation benchmark covering six distinct domains, designed to assess the robustness of evaluation metrics. In our experiments, we demonstrate that DISCODE achieves state-of-the-art performance as a reference-free evaluation metric across MCEval and four representative existing benchmarks.	Paper... Paper accepted to AAAI 2026	None
A Semantically Enhanced Generative Foundation Model Improves Pathological Image Synthesis	2025-12-15	Show The development of clinical-grade artificial intelligence in pathology is limited by the scarcity of diverse, high-quality annotated datasets. Generative models offer a potential solution but suffer from semantic instability and morphological hallucinations that compromise diagnostic reliability. To address this challenge, we introduce a Correlation-Regulated Alignment Framework for Tissue Synthesis (CRAFTS), the first generative foundation model for pathology-specific text-to-image synthesis. By leveraging a dual-stage training strategy on approximately 2.8 million image-caption pairs, CRAFTS incorporates a novel alignment mechanism that suppresses semantic drift to ensure biological accuracy. This model generates diverse pathological images spanning 30 cancer types, with quality rigorously validated by objective metrics and pathologist evaluations. Furthermore, CRAFTS-augmented datasets enhance the performance across various clinical tasks, including classification, cross-modal retrieval, self-supervised learning, and visual question answering. In addition, coupling CRAFTS with ControlNet enables precise control over tissue architecture from inputs such as nuclear segmentation masks and fluorescence images. By overcoming the critical barriers of data scarcity and privacy concerns, CRAFTS provides a limitless source of diverse, annotated histology data, effectively unlocking the creation of robust diagnostic tools for rare and complex cancer phenotypes.	67 pa... 67 pages, 9 figures, 16 tables	None
VDAWorld: World Modelling via VLM-Directed Abstraction and Simulation	2025-12-11	Show Generative video models, a leading approach to world modeling, face fundamental limitations. They often violate physical and logical rules, lack interactivity, and operate as opaque black boxes ill-suited for building structured, queryable worlds. To overcome these challenges, we propose a new paradigm focused on distilling an image caption pair into a tractable, abstract representation optimized for simulation. We introduce VDAWorld, a framework where a Vision-Language Model (VLM) acts as an intelligent agent to orchestrate this process. The VLM autonomously constructs a grounded (2D or 3D) scene representation by selecting from a suite of vision tools, and accordingly chooses a compatible physics simulator (e.g., rigid body, fluid) to act upon it. VDAWorld can then infer latent dynamics from the static scene to predict plausible future states. Our experiments show that this combination of intelligent abstraction and adaptive simulation results in a versatile world model capable of producing high quality simulations across a wide range of dynamic scenarios.	Websi... Website: https://felixomahony.github.io/vdaworld/	Code Link
Leveraging Text Guidance for Enhancing Demographic Fairness in Gender Classification	2025-12-11	Show In the quest for fairness in artificial intelligence, novel approaches to enhance it in facial image based gender classification algorithms using text guided methodologies are presented. The core methodology involves leveraging semantic information from image captions during model training to improve generalization capabilities. Two key strategies are presented: Image Text Matching (ITM) guidance and Image Text fusion. ITM guidance trains the model to discern fine grained alignments between images and texts to obtain enhanced multimodal representations. Image text fusion combines both modalities into comprehensive representations for improved fairness. Exensive experiments conducted on benchmark datasets demonstrate these approaches effectively mitigate bias and improve accuracy across gender racial groups compared to existing methods. Additionally, the unique integration of textual guidance underscores an interpretable and intuitive training paradigm for computer vision systems. By scrutinizing the extent to which semantic information reduces disparities, this research offers valuable insights into cultivating more equitable facial analysis algorithms. The proposed methodologies contribute to addressing the pivotal challenge of demographic bias in gender classification from facial images. Furthermore, this technique operates in the absence of demographic labels and is application agnostic.		None
Independent Density Estimation	2025-12-10	Show Large-scale Vision-Language models have achieved remarkable results in various domains, such as image captioning and conditioned image generation. Neverthe- less, these models still encounter difficulties in achieving human-like composi- tional generalization. In this study, we propose a new method called Independent Density Estimation (IDE) to tackle this challenge. IDE aims to learn the connec- tion between individual words in a sentence and the corresponding features in an image, enabling compositional generalization. We build two models based on the philosophy of IDE. The first one utilizes fully disentangled visual representations as input, and the second leverages a Variational Auto-Encoder to obtain partially disentangled features from raw images. Additionally, we propose an entropy- based compositional inference method to combine predictions of each word in the sentence. Our models exhibit superior generalization to unseen compositions compared to current models when evaluated on various datasets.	10 pa... 10 pages, 1 table, 4 figures	None
How a Bit Becomes a Story: Semantic Steering via Differentiable Fault Injection	2025-12-09	Show Hard-to-detect hardware bit flips, from either malicious circuitry or bugs, have already been shown to make transformers vulnerable in non-generative tasks. This work, for the first time, investigates how low-level, bitwise perturbations (fault injection) to the weights of a large language model (LLM) used for image captioning can influence the semantic meaning of its generated descriptions while preserving grammatical structure. While prior fault analysis methods have shown that flipping a few bits can crash classifiers or degrade accuracy, these approaches overlook the semantic and linguistic dimensions of generative systems. In image captioning models, a single flipped bit might subtly alter how visual features map to words, shifting the entire narrative an AI tells about the world. We hypothesize that such semantic drifts are not random but differentiably estimable. That is, the model's own gradients can predict which bits, if perturbed, will most strongly influence meaning while leaving syntax and fluency intact. We design a differentiable fault analysis framework, BLADE (Bit-level Fault Analysis via Differentiable Estimation), that uses gradient-based sensitivity estimation to locate semantically critical bits and then refines their selection through a caption-level semantic-fluency objective. Our goal is not merely to corrupt captions, but to understand how meaning itself is encoded, distributed, and alterable at the bit level, revealing that even imperceptible low-level changes can steer the high-level semantics of generative vision-language models. It also opens pathways for robustness testing, adversarial defense, and explainable AI, by exposing how structured bit-level faults can reshape a model's semantic output.		None
Explaining the Unseen: Multimodal Vision-Language Reasoning for Situational Awareness in Underground Mining Disasters	2025-12-09	Show Underground mining disasters produce pervasive darkness, dust, and collapses that obscure vision and make situational awareness difficult for humans and conventional systems. To address this, we propose MDSE, Multimodal Disaster Situation Explainer, a novel vision-language framework that automatically generates detailed textual explanations of post-disaster underground scenes. MDSE has three-fold innovations: (i) Context-Aware Cross-Attention for robust alignment of visual and textual features even under severe degradation; (ii) Segmentation-aware dual pathway visual encoding that fuses global and region-specific embeddings; and (iii) Resource-Efficient Transformer-Based Language Model for expressive caption generation with minimal compute cost. To support this task, we present the Underground Mine Disaster (UMD) dataset--the first image-caption corpus of real underground disaster scenes--enabling rigorous training and evaluation. Extensive experiments on UMD and related benchmarks show that MDSE substantially outperforms state-of-the-art captioning models, producing more accurate and contextually relevant descriptions that capture crucial details in obscured environments, improving situational awareness for underground emergency response. The code is at https://github.com/mizanJewel/Multimodal-Disaster-Situation-Explainer.		Code Link
Siamese-Driven Optimization for Low-Resolution Image Latent Embedding in Image Captioning	2025-12-09	Show Image captioning is essential in many fields including assisting visually impaired individuals, improving content management systems, and enhancing human-computer interaction. However, a recent challenge in this domain is dealing with low-resolution image (LRI). While performance can be improved by using larger models like transformers for encoding, these models are typically heavyweight, demanding significant computational resources and memory, leading to challenges in retraining. To address this, the proposed SOLI (Siamese-Driven Optimization for Low-Resolution Image Latent Embedding in Image Captioning) approach presents a solution specifically designed for lightweight, low-resolution images captioning. It employs a Siamese network architecture to optimize latent embeddings, enhancing the efficiency and accuracy of the image-to-text translation process. By focusing on a dual-pathway neural network structure, SOLI minimizes computational overhead without sacrificing performance, making it an ideal choice for training on resource-constrained scenarios.	6 pages	None
Leveraging Machine Learning and Large Language Models for Automated Image Clustering and Description in Legal Discovery	2025-12-08	Show The rapid increase in digital image creation and retention presents substantial challenges during legal discovery, digital archive, and content management. Corporations and legal teams must organize, analyze, and extract meaningful insights from large image collections under strict time pressures, making manual review impractical and costly. These demands have intensified interest in automated methods that can efficiently organize and describe large-scale image datasets. This paper presents a systematic investigation of automated cluster description generation through the integration of image clustering, image captioning, and large language models (LLMs). We apply K-means clustering to group images into 20 visually coherent clusters and generate base captions using the Azure AI Vision API. We then evaluate three critical dimensions of the cluster description process: (1) image sampling strategies, comparing random, centroid-based, stratified, hybrid, and density-based sampling against using all cluster images; (2) prompting techniques, contrasting standard prompting with chain-of-thought prompting; and (3) description generation methods, comparing LLM-based generation with traditional TF-IDF and template-based approaches. We assess description quality using semantic similarity and coverage metrics. Results show that strategic sampling with 20 images per cluster performs comparably to exhaustive inclusion while significantly reducing computational cost, with only stratified sampling showing modest degradation. LLM-based methods consistently outperform TF-IDF baselines, and standard prompts outperform chain-of-thought prompts for this task. These findings provide practical guidance for deploying scalable, accurate cluster description systems that support high-volume workflows in legal discovery and other domains requiring automated organization of large image collections.		None
Relational Visual Similarity	2025-12-08	Show Humans do not just see attribute similarity -- we also see relational similarity. An apple is like a peach because both are reddish fruit, but the Earth is also like a peach: its crust, mantle, and core correspond to the peach's skin, flesh, and pit. This ability to perceive and recognize relational similarity, is arguable by cognitive scientist to be what distinguishes humans from other species. Yet, all widely used visual similarity metrics today (e.g., LPIPS, CLIP, DINO) focus solely on perceptual attribute similarity and fail to capture the rich, often surprising relational similarities that humans perceive. How can we go beyond the visible content of an image to capture its relational properties? How can we bring images with the same relational logic closer together in representation space? To answer these questions, we first formulate relational image similarity as a measurable problem: two images are relationally similar when their internal relations or functions among visual elements correspond, even if their visual attributes differ. We then curate 114k image-caption dataset in which the captions are anonymized -- describing the underlying relational logic of the scene rather than its surface content. Using this dataset, we finetune a Vision-Language model to measure the relational similarity between images. This model serves as the first step toward connecting images by their underlying relational structure rather than their visible appearance. Our study shows that while relational similarity has a lot of real-world applications, existing image similarity models fail to capture it -- revealing a critical gap in visual computing.	Proje... Project page, data, and code: https://thaoshibe.github.io/relsim	Code Link
MMRPT: MultiModal Reinforcement Pre-Training via Masked Vision-Dependent Reasoning	2025-12-08	Show Multimodal pre-training remains constrained by the descriptive bias of image-caption pairs, leading models to favor surface linguistic cues over grounded visual understanding. We introduce MMRPT, a masked multimodal reinforcement pre-training framework that strengthens visual reasoning in MLLMs. We are the first to incorporate reinforcement learning directly into the pre-training of large vision-language models, enabling learning signals that reward visual grounding rather than caption imitation. MMRPT constructs masked multimodal data by estimating sentence-level visual dependency via attention over visual tokens and masking highly vision-dependent segments; the model reconstructs these spans through vision-grounded reasoning guided by a semantic-visual reward. Experiments show consistent zero-shot gains across diverse benchmarks and substantially improved robustness under supervised fine-tuning, demonstrating that reinforcement-driven masked reasoning provides a more reliable and generalizable pre-training objective for multimodal models.	7 pages, 1 figures	None
LLM as a Neural Architect: Controlled Generation of Image Captioning Models Under Strict API Contracts	2025-12-07	Show Neural architecture search (NAS) traditionally requires significant human expertise or automated trial-and-error to design deep learning models. We present NN-Caption, an LLM-guided neural architecture search pipeline that generates runnable image-captioning models by composing CNN encoders from LEMUR's classification backbones with sequence decoders (LSTM/GRU/Transformer) under a strict Net API. Using DeepSeek-R1-0528-Qwen3-8B as the primary generator, we present the prompt template and examples of generated architectures. We evaluate on MS COCO with BLEU-4. The LLM generated dozens of captioning models, with over half successfully trained and producing meaningful captions. We analyse the outcomes of using different numbers of input model snippets (5 vs. 10) in the prompt, finding a slight drop in success rate when providing more candidate components. We also report training dynamics (caption accuracy vs. epochs) and the highest BLEU-4 attained. Our results highlight the promise of LLM-guided NAS: the LLM not only proposes architectures but also suggests hyperparameters and training practices. We identify the challenges encountered (e.g., code hallucinations or API compliance issues) and detail how prompt rules and iterative code fixes addressed them. This work presents a pipeline that integrates prompt-based code generation with automatic evaluation, and adds dozens of novel captioning models to the open LEMUR dataset to facilitate reproducible benchmarking and downstream AutoML research.		None
AgriGPT-VL: Agricultural Vision-Language Understanding Suite	2025-12-07	Show Despite rapid advances in multimodal large language models, agricultural applications remain constrained by the scarcity of domain-tailored models, curated vision-language corpora, and rigorous evaluation. To address these challenges, we present the AgriGPT-VL Suite, a unified multimodal framework for agriculture. Our contributions are threefold. First, we introduce Agri-3M-VL, the largest vision-language corpus for agriculture to our knowledge, curated by a scalable multi-agent data generator; it comprises 1M image-caption pairs, 2M image-grounded VQA pairs, 50K expert-level VQA instances, and 15K GRPO reinforcement learning samples. Second, we develop AgriGPT-VL, an agriculture-specialized vision-language model trained via a progressive curriculum of textual grounding, multimodal shallow/deep alignment, and GRPO refinement. This method achieves strong multimodal reasoning while preserving text-only capability. Third, we establish AgriBench-VL-4K, a compact yet challenging evaluation suite with open-ended and image-grounded questions, paired with multi-metric evaluation and an LLM-as-a-judge framework. Experiments show that AgriGPT-VL outperforms leading general-purpose VLMs on AgriBench-VL-4K, achieving higher pairwise win rates in the LLM-as-a-judge evaluation. Meanwhile, it remains competitive on the text-only AgriBench-13K with no noticeable degradation of language ability. Ablation studies further confirm consistent gains from our alignment and GRPO refinement stages. We will open source all of the resources to support reproducible research and deployment in low-resource agricultural settings.		None
MM-SeR: Multimodal Self-Refinement for Lightweight Image Captioning	2025-12-07	Show Systems such as video chatbots and navigation robots often depend on streaming image captioning to interpret visual inputs. Existing approaches typically employ large multimodal language models (MLLMs) for this purpose, but their substantial computational cost hinders practical application. This limitation motivates our development of a lightweight captioning model. Our investigation begins by replacing the large-scale language component in MLLMs with a compact 125M-parameter model. Surprisingly, this compact model, despite a 93x reduction in size, achieves comparable performance to MLLMs, suggesting that factual image captioning does not significantly require the complex reasoning abilities of LLMs. Despite this promising result, our lightweight model still lacks reliability. To address this, we draw inspiration from the human visual process: perceiving a global and coarse understanding of the scene before attending to finer details. Accordingly, we propose a multimodal self-refinement framework that guides the model to utilize features from salient regions, identified by referencing the previous coarse caption, and to produce a refined description. Experimental results demonstrate the superiority of our model in both single-sentence and detailed captioning, extending even to long-range video QA tasks.	Proje... Project page: https://sites.google.com/view/junha/mm-ser	None
Open-PMC-18M: A High-Fidelity Large Scale Medical Dataset for Multimodal Representation Learning	2025-12-05	Show In biomedical vision-language modeling, datasets are typically mined from scientific literature, pairing compound figures with captions that are short, context-dependent, and oftern partially informative. Prior work on subfigure extraction has been limited in both dataset size and generalizability. In addition, no existing effort has incorporated rich medical context in image-text pairs. We revisit data curation as a foundational component of effective biomedical representation learning. Our data curation process integrates transformer-based subfigure detection, subcaption extraction, and contextual text enrichment derived from inline references. Our subfigure extraction model, trained on a corpus of 500,000 compound figures, achieves state-of-the-art performance on real and synthetic benchmarks. Using this process, we curate and release Open-PMC-18M, a large-scale high-fidelity biomedical dataset comprising 18 million image-text pairs, spanning radiology, microscopy, and visible light photography. We train vision-language models on our dataset and perform extensive evaluation on 6 retrieval and 19 zero-shot classification tasks across three major modalities. The models trained on our dataset set a new state-of-the-art results in medical representation learning. We release our dataset, models, and code to support reproducible benchmarks and further study into biomedical vision-language modeling and representation learning.	21 pages	None
A quantitative analysis of semantic information in deep representations of text and images	2025-12-05	Show Deep neural networks are known to develop similar representations for semantically related data, even when they belong to different domains, such as an image and its description, or the same text in different languages. We present a method for quantitatively investigating this phenomenon by measuring the relative information content of the representations of semantically related data and probing how it is encoded into multiple tokens of large language models (LLMs) and vision transformers. Looking first at how LLMs process pairs of translated sentences, we identify inner ``semantic'' layers containing the most language-transferable information. We find moreover that, on these layers, a larger LLM (DeepSeek-V3) extracts significantly more general information than a smaller one (Llama3.1-8B). Semantic information of English text is spread across many tokens and it is characterized by long-distance correlations between tokens and by a causal left-to-right (i.e., past-future) asymmetry. We also identify layers encoding semantic information within visual transformers. We show that caption representations in the semantic layers of LLMs predict visual representations of the corresponding images. We observe significant and model-dependent information asymmetries between image and text representations.		None
Mitigating Object and Action Hallucinations in Multimodal LLMs via Self-Augmented Contrastive Alignment	2025-12-04	Show Recent advancement in multimodal LLMs (MLLMs) has demonstrated their remarkable capability to generate descriptive captions for input videos. However, these models suffer from factual inaccuracies in the generated descriptions, causing severe hallucination issues. While prior works have explored alleviating hallucinations for static images, jointly mitigating visual object and temporal action hallucinations for dynamic videos remains a challenging and unsolved task. To tackle this challenge, we propose a Self-Augmented Contrastive Alignment (SANTA) framework for enabling object and action faithfulness by exempting the spurious correlations and enforcing the emphasis on visual facts. SANTA employs a hallucinative self-augmentation scheme to identify the potential hallucinations that lie in the MLLM and transform the original captions to the contrasted negatives. Furthermore, we develop a tracklet-phrase contrastive alignment to match the regional objects and relation-guided actions with their corresponding visual and temporal phrases. Extensive experiments demonstrate that SANTA outperforms existing methods in alleviating object and action hallucinations, yielding superior performance on the hallucination examination benchmarks.	IEEE/... IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) 2026. Project page: https://kpc0810.github.io/santa/	Code Link
Text-Only Training for Image Captioning with Retrieval Augmentation and Modality Gap Correction	2025-12-03	Show Image captioning has drawn considerable attention from the natural language processing and computer vision fields. Aiming to reduce the reliance on curated data, several studies have explored image captioning without any humanly-annotated image-text pairs for training, although existing methods are still outperformed by fully supervised approaches. This paper proposes TOMCap, i.e., an improved text-only training method that performs captioning without the need for aligned image-caption pairs. The method is based on prompting a pre-trained language model decoder with information derived from a CLIP representation, after undergoing a process to reduce the modality gap. We specifically tested the combined use of retrieved examples of captions, and latent vector representations, to guide the generation process. Through extensive experiments, we show that TOMCap outperforms other training-free and text-only methods. We also analyze the impact of different choices regarding the configuration of the retrieval-augmentation and modality gap reduction components.	Submi... Submitted to CVPR 2026	None
GT23D-Bench: A Comprehensive General Text-to-3D Generation Benchmark	2025-12-03	Show Text-to-3D (T23D) generation has emerged as a crucial visual generation task, aiming at synthesizing 3D content from textual descriptions. Studies of this task are currently shifting from per-scene T23D, which requires optimization of the model for every content generated, to General T23D (GT23D), which requires only one pre-trained model to generate different content without re-optimization, for more generalized and efficient 3D generation. Despite notable advancements, GT23D is severely bottlenecked by two interconnected challenges: the lack of high-quality, large-scale training data and the prevalence of evaluation metrics that overlook intrinsic 3D properties. Existing datasets often suffer from incomplete annotations, noisy organization, and inconsistent quality, while current evaluations rely heavily on 2D image-text similarity or scoring, failing to thoroughly assess 3D geometric integrity and semantic relevance. To address these fundamental gaps, we introduce GT23D-Bench, the first comprehensive benchmark specifically designed for GT23D training and evaluation. We first construct a high-quality dataset of 400K 3D assets, featuring diverse visual annotations (70M+ visual samples) and multi-granularity hierarchical captions (1M+ descriptions) to foster robust semantic learning. Second, we propose a comprehensive evaluation suite with 10 metrics assessing both text-3D alignment and 3D visual quality at multiple levels. Crucially, we demonstrate through rigorous experiments that our proposed metrics exhibit significantly higher correlation with human judgment compared to existing methods. Our in-depth analysis of eight leading GT23D models using this benchmark provides the community with critical insights into current model capabilities and their shared failure modes. GT23D-Bench will be publicly available to facilitate rigorous and reproducible research.		None
SATORI-R1: Incentivizing Multimodal Reasoning through Explicit Visual Anchoring	2025-12-03	Show DeepSeek-R1 has demonstrated powerful reasoning capabilities in the text domain through stable reinforcement learning (RL). Recently, in the multimodal domain, works have begun to directly apply RL to generate R1-like free-form reasoning for Visual Question Answering (VQA) tasks. However, multimodal tasks share an intrinsically different nature from textual tasks, which heavily rely on the understanding of the input image to solve the problem. Therefore, such free-form reasoning faces two critical limitations in the VQA task: (1) Extended reasoning chains diffuse visual focus away from task-critical regions, degrading answer accuracy. (2) Unverifiable intermediate steps amplify policy-gradient variance and computational costs overhead. To address these issues, in this paper, we introduce SATORI ($\textbf{S}patially$ $\textbf{A}nchored$ $\textbf{T}ask$ $\textbf{O}ptimization$ with $\textbf{R}e\textbf{I}nforcement$ Learning), which decomposes VQA into three verifiable stages, including global image captioning, region localization, and answer prediction, each supplying explicit reward signals. Furthermore, we also introduce VQA-Verify, a 12k dataset annotated with answer-aligned captions and bounding-boxes to facilitate training. Experiments demonstrate consistent performance improvements across seven VQA benchmarks, achieving up to $15.7%$ improvement in accuracy in accuracy compared to the R1-like baseline. Our analysis of the attention map confirms enhanced focus on critical regions, which brings improvements in accuracy. Our code is available at https://github.com/justairr/SATORI-R1.	21 pages, 8 figures	Code Link
Multi-Aspect Knowledge-Enhanced Medical Vision-Language Pretraining with Multi-Agent Data Generation	2025-12-03	Show Vision-language pretraining (VLP) has emerged as a powerful paradigm in medical image analysis, enabling representation learning from large-scale image-text pairs without relying on expensive manual annotations. However, existing methods often struggle with the noise inherent in web-collected data and the complexity of unstructured long medical texts. To address these challenges, we propose a novel VLP framework integrating a Multi-Agent data GENeration (MAGEN) system and Ontology-based Multi-Aspect Knowledge-Enhanced (O-MAKE) pretraining. First, MAGEN enhances data quality by synthesizing knowledge-enriched descriptions via a foundation model-assisted captioning and retrieval-based verification pipeline. Second, O-MAKE addresses the difficulty of learning from long, unstructured texts by decomposing them into distinct knowledge aspects. This facilitates fine-grained alignment at both global and patch levels, while explicitly modeling medical concept relationships through ontology-guided mechanisms. We validate our framework in the field of dermatology, where comprehensive experiments demonstrate the effectiveness of each component. Our approach achieves state-of-the-art zero-shot performance on disease classification and cross-modal retrieval tasks across eight datasets. Our code and the augmented dataset Derm1M-AgentAug, comprising over 400k skin-image-text pairs, will be released at https://github.com/SiyuanYan1/Derm1M.	10 pa... 10 pages. Under Review	Code Link
ViDiC: Video Difference Captioning	2025-12-03	Show Understanding visual differences between dynamic scenes requires the comparative perception of compositional, spatial, and temporal changes--a capability that remains underexplored in existing vision-language systems. While prior work on Image Difference Captioning (IDC) has enabled models to describe semantic changes between static images, these approaches fail to capture motion continuity, event evolution, or editing consistency over time. We introduce the ViDiC (Video Difference Captioning) task and its corresponding ViDiC-1K dataset, designed to evaluate the ability of Multimodal Large Language Models (MLLMs) to provide fine-grained descriptions of similarities and differences between video pairs. ViDiC-1K comprises 1,000 curated video pairs annotated with over 4,000 comparative checklist items, covering seven categories: subject, style, background, cinematography, motion, location, and playback techniques. To ensure reliable evaluation, we propose a dual-checklist framework that measures the accuracy of similarity and difference separately, based on the LLM-as-a-Judge protocol. Experiments on nineteen representative multimodal models reveal a significant performance gap in their comparative description and difference perception abilities. We hope ViDiC-1K can be a challenging benchmark that lays a solid foundation for advancing video understanding, edit awareness, and comparative reasoning in multimodal intelligence.		None
OneThinker: All-in-one Reasoning Model for Image and Video	2025-12-02	Show Reinforcement learning (RL) has recently achieved remarkable success in eliciting visual reasoning within Multimodal Large Language Models (MLLMs). However, existing approaches typically train separate models for different tasks and treat image and video reasoning as disjoint domains. This results in limited scalability toward a multimodal reasoning generalist, which restricts practical versatility and hinders potential knowledge sharing across tasks and modalities. To this end, we propose OneThinker, an all-in-one reasoning model that unifies image and video understanding across diverse fundamental visual tasks, including question answering, captioning, spatial and temporal grounding, tracking, and segmentation. To achieve this, we construct the OneThinker-600k training corpus covering all these tasks and employ commercial models for CoT annotation, resulting in OneThinker-SFT-340k for SFT cold start. Furthermore, we propose EMA-GRPO to handle reward heterogeneity in multi-task RL by tracking task-wise moving averages of reward standard deviations for balanced optimization. Extensive experiments on diverse visual benchmarks show that OneThinker delivers strong performance on 31 benchmarks, across 10 fundamental visual understanding tasks. Moreover, it exhibits effective knowledge transfer between certain tasks and preliminary zero-shot generalization ability, marking a step toward a unified multimodal reasoning generalist. All code, model, and data are released.	Proje... Project page: https://github.com/tulerfeng/OneThinker	Code Link
MultiShotMaster: A Controllable Multi-Shot Video Generation Framework	2025-12-02	Show Current video generation techniques excel at single-shot clips but struggle to produce narrative multi-shot videos, which require flexible shot arrangement, coherent narrative, and controllability beyond text prompts. To tackle these challenges, we propose MultiShotMaster, a framework for highly controllable multi-shot video generation. We extend a pretrained single-shot model by integrating two novel variants of RoPE. First, we introduce Multi-Shot Narrative RoPE, which applies explicit phase shift at shot transitions, enabling flexible shot arrangement while preserving the temporal narrative order. Second, we design Spatiotemporal Position-Aware RoPE to incorporate reference tokens and grounding signals, enabling spatiotemporal-grounded reference injection. In addition, to overcome data scarcity, we establish an automated data annotation pipeline to extract multi-shot videos, captions, cross-shot grounding signals and reference images. Our framework leverages the intrinsic architectural properties to support multi-shot video generation, featuring text-driven inter-shot consistency, customized subject with motion control, and background-driven customized scene. Both shot count and duration are flexibly configurable. Extensive experiments demonstrate the superior performance and outstanding controllability of our framework.	Proje... Project Page: https://qinghew.github.io/MultiShotMaster	Code Link
GeoDiT: A Diffusion-based Vision-Language Model for Geospatial Understanding	2025-12-02	Show Autoregressive models are structurally misaligned with the inherently parallel nature of geospatial understanding, forcing a rigid sequential narrative onto scenes and fundamentally hindering the generation of structured and coherent outputs. We challenge this paradigm by reframing geospatial generation as a parallel refinement process, enabling a holistic, coarse-to-fine synthesis that resolves all semantic elements simultaneously. To operationalize this, we introduce GeoDiT, the first diffusion-based vision-language model tailored for the geospatial domain. Extensive experiments demonstrate that GeoDiT establishes a new state-of-the-art on benchmarks requiring structured, object-centric outputs. It achieves significant gains in image captioning, visual grounding, and multi-object detection, precisely the tasks where autoregressive models falter. Our work validates that aligning the generative process with the data's intrinsic structure is key to unlocking superior performance in complex geospatial analysis.		None
Rainbow Noise: Stress-Testing Multimodal Harmful-Meme Detectors on LGBTQ Content	2025-12-02	Show Hateful memes aimed at LGBTQ,+ communities often evade detection by tweaking either the caption, the image, or both. We build the first robustness benchmark for this setting, pairing four realistic caption attacks with three canonical image corruptions and testing all combinations on the PrideMM dataset. Two state-of-the-art detectors, MemeCLIP and MemeBLIP2, serve as case studies, and we introduce a lightweight \textbf{Text Denoising Adapter (TDA)} to enhance the latter's resilience. Across the grid, MemeCLIP degrades more gently, while MemeBLIP2 is particularly sensitive to the caption edits that disrupt its language processing. However, the addition of the TDA not only remedies this weakness but makes MemeBLIP2 the most robust model overall. Ablations reveal that all systems lean heavily on text, but architectural choices and pre-training data significantly impact robustness. Our benchmark exposes where current multimodal safety models crack and demonstrates that targeted, lightweight modules like the TDA offer a powerful path towards stronger defences.	14 pages, 1 figure	None
SGDiff: Scene Graph Guided Diffusion Model for Image Collaborative SegCaptioning	2025-12-01	Show Controllable image semantic understanding tasks, such as captioning or segmentation, necessitate users to input a prompt (e.g., text or bounding boxes) to predict a unique outcome, presenting challenges such as high-cost prompt input or limited information output. This paper introduces a new task ``Image Collaborative Segmentation and Captioning'' (SegCaptioning), which aims to translate a straightforward prompt, like a bounding box around an object, into diverse semantic interpretations represented by (caption, masks) pairs, allowing flexible result selection by users. This task poses significant challenges, including accurately capturing a user's intention from a minimal prompt while simultaneously predicting multiple semantically aligned caption words and masks. Technically, we propose a novel Scene Graph Guided Diffusion Model that leverages structured scene graph features for correlated mask-caption prediction. Initially, we introduce a Prompt-Centric Scene Graph Adaptor to map a user's prompt to a scene graph, effectively capturing his intention. Subsequently, we employ a diffusion process incorporating a Scene Graph Guided Bimodal Transformer to predict correlated caption-mask pairs by uncovering intricate correlations between them. To ensure accurate alignment, we design a Multi-Entities Contrastive Learning loss to explicitly align visual and textual entities by considering inter-modal similarity, resulting in well-aligned caption-mask pairs. Extensive experiments conducted on two datasets demonstrate that SGDiff achieves superior performance in SegCaptioning, yielding promising results for both captioning and segmentation tasks with minimal prompt input.	Accept by AAAI-2025	None
InternVideo-Next: Towards General Video Foundation Models without Video-Text Supervision	2025-12-01	Show Large-scale video-text pretraining achieves strong performance but depends on noisy, synthetic captions with limited semantic coverage, often overlooking implicit world knowledge such as object motion, 3D geometry, and physical cues. In contrast, masked video modeling (MVM) directly exploits spatiotemporal structures but trails text-supervised methods on general tasks. We find this gap arises from overlooked architectural issues: pixel-level reconstruction struggles with convergence and its low-level requirement often conflicts with semantics, while latent prediction often encourages shortcut learning. To address these, we disentangle the traditional encoder-decoder design into an Encoder-Predictor-Decoder (EPD) framework, where the predictor acts as a latent world model, and propose InternVideo-Next, a two-stage pretraining scheme that builds a semantically consistent yet detail-preserving latent space for this world model. First, conventional linear decoder in pixel MVM enforces the predictor output latent to be linearly projected to, thus separable in pixel space, causing the conflict with semantic abstraction. Our Stage 1 proposes a conditional diffusion decoder and injects reliable image-level semantic priors to enhance semantics and convergence, thus bridging pixel-level fidelity with high-level semantic abstraction. Stage 2 further learns world knowledge by predicting frozen Stage 1 targets within this space, mitigating shortcut learning. Trained on public, unlabeled videos, InternVideo-Next achieves state-of-the-art results across benchmarks and provides a scalable path toward general video representation learning.		None
Hierarchical Semantic Alignment for Image Clustering	2025-11-30	Show Image clustering is a classic problem in computer vision, which categorizes images into different groups. Recent studies utilize nouns as external semantic knowledge to improve clus- tering performance. However, these methods often overlook the inherent ambiguity of nouns, which can distort semantic representations and degrade clustering quality. To address this issue, we propose a hierarChical semAntic alignmEnt method for image clustering, dubbed CAE, which improves cluster- ing performance in a training-free manner. In our approach, we incorporate two complementary types of textual seman- tics: caption-level descriptions, which convey fine-grained attributes of image content, and noun-level concepts, which represent high-level object categories. We first select relevant nouns from WordNet and descriptions from caption datasets to construct a semantic space aligned with image features. Then, we align image features with selected nouns and captions via optimal transport to obtain a more discriminative semantic space. Finally, we combine the enhanced semantic and image features to perform clustering. Extensive experiments across 8 datasets demonstrate the effectiveness of our method, notably surpassing the state-of-the-art training-free approach with a 4.2% improvement in accuracy and a 2.9% improvement in adjusted rand index (ARI) on the ImageNet-1K dataset.	AAAI 2026	None
Multilingual Training-Free Remote Sensing Image Captioning	2025-11-30	Show Remote sensing image captioning has advanced rapidly through encoder--decoder models, although the reliance on large annotated datasets and the focus on English restricts global applicability. To address these limitations, we propose the first training-free multilingual approach, based on retrieval-augmented prompting. For a given aerial image, we employ a domain-adapted SigLIP2 encoder to retrieve related captions and few-shot examples from a datastore, which are then provided to a language model. We explore two variants: an image-blind setup, where a multilingual Large Language Model (LLM) generates the caption from textual prompts alone, and an image-aware setup, where a Vision--Language Model (VLM) jointly processes the prompt and the input image. To improve the coherence of the retrieved content, we introduce a graph-based re-ranking strategy using PageRank on a graph of images and captions. Experiments on four benchmark datasets across ten languages demonstrate that our approach is competitive with fully supervised English-only systems and generalizes to other languages. Results also highlight the importance of re-ranking with PageRank, yielding up to 35% improvements in performance metrics. Additionally, it was observed that while VLMs tend to generate visually grounded but lexically diverse captions, LLMs can achieve stronger BLEU and CIDEr scores. Lastly, directly generating captions in the target language consistently outperforms other translation-based strategies. Overall, our work delivers one of the first systematic evaluations of multilingual, training-free captioning for remote sensing imagery, advancing toward more inclusive and scalable multimodal Earth observation systems.		None
BioPro: On Difference-Aware Gender Fairness for Vision-Language Models	2025-11-30	Show Vision-Language Models (VLMs) inherit significant social biases from their training data, notably in gender representation. Current fairness interventions often adopt a difference-unaware perspective that enforces uniform treatment across demographic groups. These approaches, however, fail to distinguish between contexts where neutrality is required and those where group-specific attributes are legitimate and must be preserved. Building upon recent advances in difference-aware fairness for text-only models, we extend this concept to the multimodal domain and formalize the problem of difference-aware gender fairness for image captioning and text-to-image generation. We advocate for selective debiasing, which aims to mitigate unwanted bias in neutral contexts while preserving valid distinctions in explicit ones. To achieve this, we propose BioPro (Bias Orthogonal Projection), an entirely training-free framework. BioPro identifies a low-dimensional gender-variation subspace through counterfactual embeddings and applies projection to selectively neutralize gender-related information. Experiments show that BioPro effectively reduces gender bias in neutral cases while maintaining gender faithfulness in explicit ones, thus providing a promising direction toward achieving selective fairness in VLMs. Beyond gender bias, we further demonstrate that BioPro can effectively generalize to continuous bias variables, such as scene brightness, highlighting its broader applicability.		None
DEJIMA: A Novel Large-scale Japanese Dataset for Image Captioning and Visual Question Answering	2025-11-30	Show This work addresses the scarcity of high-quality, large-scale resources for Japanese Vision-and-Language (V&L) modeling. We present a scalable and reproducible pipeline that integrates large-scale web collection with rigorous filtering/deduplication, object-detection-driven evidence extraction, and Large Language Model (LLM)-based refinement under grounding constraints. Using this pipeline, we build two resources: an image-caption dataset (DEJIMA-Cap) and a VQA dataset (DEJIMA-VQA), each containing 3.88M image-text pairs, far exceeding the size of existing Japanese V&L datasets. Human evaluations demonstrate that DEJIMA achieves substantially higher Japaneseness and linguistic naturalness than datasets constructed via translation or manual annotation, while maintaining factual correctness at a level comparable to human-annotated corpora. Quantitative analyses of image feature distributions further confirm that DEJIMA broadly covers diverse visual domains characteristic of Japan, complementing its linguistic and cultural representativeness. Models trained on DEJIMA exhibit consistent improvements across multiple Japanese multimodal benchmarks, confirming that culturally grounded, large-scale resources play a key role in enhancing model performance. All data sources and modules in our pipeline are licensed for commercial use, and we publicly release the resulting dataset and metadata to encourage further research and industrial applications in Japanese V&L modeling.		None
CLIP-Free, Label-Free, Zero-Shot Concept Bottleneck Models	2025-11-28	Show Concept Bottleneck Models (CBMs) map dense, high-dimensional feature representations into a set of human-interpretable concepts which are then combined linearly to make a prediction. However, modern CBMs rely on the CLIP model to establish a mapping from dense feature representations to textual concepts, and it remains unclear how to design CBMs for models other than CLIP. Methods that do not use CLIP instead require manual, labor intensive annotation to associate feature representations with concepts. Furthermore, all CBMs necessitate training a linear classifier to map the extracted concepts to class labels. In this work, we lift all three limitations simultaneously by proposing a method that converts any frozen visual classifier into a CBM without requiring image-concept labels (label-free), without relying on the CLIP model (CLIP-free), and by deriving the linear classifier in a zero-shot manner. Our method is formulated by aligning the original classifier's distribution (over discrete class indices) with its corresponding vision-language counterpart distribution derived from textual class names, while preserving the classifier's performance. The approach requires no ground-truth image-class annotations, and is highly data-efficient and preserves the classifier's reasoning process. Applied and tested on over 40 visual classifiers, our resulting CLIP-free, zero-shot CBM sets a new state of the art, surpassing even supervised CLIP-based CBMs. Finally, we also show that our method can be used for zero-shot image captioning, outperforming existing methods based on CLIP, and achieving state of the art results.		None
DenseScan: Advancing 3D Scene Understanding with 2D Dense Annotation	2025-11-28	Show 3D understanding is a key capability for real-world AI assistance. High-quality data plays an important role in driving the development of the 3D understanding community. Current 3D scene understanding datasets often provide geometric and instance-level information, yet they lack the rich semantic annotations necessary for nuanced visual-language tasks.In this work, we introduce DenseScan, a novel dataset with detailed multi-level descriptions generated by an automated pipeline leveraging multi-view 2D images and multimodal large language models (MLLMs). Our approach enables dense captioning of scene elements, ensuring comprehensive object-level descriptions that capture context-sensitive details. Furthermore, we extend these annotations through scenario-based question generation, producing high-level queries that integrate object properties, spatial relationships, and scene context. By coupling geometric detail with semantic richness, DenseScan broadens the range of downstream tasks, from detailed visual-language navigation to interactive question answering. Experimental results demonstrate that our method significantly enhances object-level understanding and question-answering performance in 3D environments compared to traditional annotation pipelines. We release both the annotated dataset and our annotation pipeline to facilitate future research and applications in robotics, augmented reality, and beyond. Through DenseScan, we aim to catalyze new avenues in 3D scene understanding, allowing researchers and practitioners to tackle the complexities of real-world environments with richer, more contextually aware annotations.	Works... Workshop on Space in Vision, Language, and Embodied AI at NeurIPS 2025	None
Optimizing Multimodal Language Models through Attention-based Interpretability	2025-11-28	Show Modern large language models become multimodal, analyzing various data formats like text and images. While fine-tuning is effective for adapting these multimodal language models (MLMs) to downstream tasks, full fine-tuning is computationally expensive. Parameter-Efficient Fine-Tuning (PEFT) methods address this by training only a small portion of model weights. However, MLMs are difficult to interpret, making it challenging to identify which components are most effective for training to balance efficiency and performance. We propose an attention-based interpretability method for MLMs by analyzing attention scores relative to image tokens. The core idea is to identify attention heads that focus on image key objects. We utilize this information to select optimal model components for PEFT in multimodal models. Our contributions include a method for identifying attention heads associated with image key objects, its application to PEFT for image captioning, and the creation of a new dataset containing images, key object masks, and their textual descriptions. We conducted experiments on MLMs with 2-3 billion parameters to validate the method's effectiveness. By calculating Head Impact (HI) scores we quantify an attention head's focus on key objects, indicating its significance in image understanding. Our fine-tuning experiments demonstrate that adapting layers with the highest HI scores leads to the most significant shifts in metrics compared to pre-trained, randomly selected, or lowest-HI-score layers. This indicates that fine-tuning a small percentage (around 0.01%) of parameters in these crucial layers can substantially influence image understanding capabilities.	Accep... Accepted for ICAI-2025 conference	None
Analyzing Image Beyond Visual Aspect: Image Emotion Classification via Multiple-Affective Captioning	2025-11-28	Show Image emotion classification (IEC) is a longstanding research field that has received increasing attention with the rapid progress of deep learning. Although recent advances have leveraged the knowledge encoded in pre-trained visual models, their effectiveness is constrained by the "affective gap" , limits the applicability of pre-training knowledge for IEC tasks. It has been demonstrated in psychology that language exhibits high variability, encompasses diverse and abundant information, and can effectively eliminate the "affective gap". Inspired by this, we propose a novel Affective Captioning for Image Emotion Classification (ACIEC) to classify image emotion based on pure texts, which effectively capture the affective information in the image. In our method, a hierarchical multi-level contrastive loss is designed for detecting emotional concepts from images, while an emotional attribute chain-of-thought reasoning is proposed to generate affective sentences. Then, a pre-trained language model is leveraged to synthesize emotional concepts and affective sentences to conduct IEC. Additionally, a contrastive loss based on semantic similarity sampling is designed to solve the problem of large intra-class differences and small inter-class differences in affective datasets. Moreover, we also take the images with embedded texts into consideration, which were ignored by previous studies. Extensive experiments illustrate that our method can effectively bridge the affective gap and achieve superior results on multiple benchmarks.		None
Leveraging Textual Compositional Reasoning for Robust Change Captioning	2025-11-28	Show Change captioning aims to describe changes between a pair of images. However, existing works rely on visual features alone, which often fail to capture subtle but meaningful changes because they lack the ability to represent explicitly structured information such as object relationships and compositional semantics. To alleviate this, we present CORTEX (COmpositional Reasoning-aware TEXt-guided), a novel framework that integrates complementary textual cues to enhance change understanding. In addition to capturing cues from pixel-level differences, CORTEX utilizes scene-level textual knowledge provided by Vision Language Models (VLMs) to extract richer image text signals that reveal underlying compositional reasoning. CORTEX consists of three key modules: (i) an Image-level Change Detector that identifies low-level visual differences between paired images, (ii) a Reasoning-aware Text Extraction (RTE) module that use VLMs to generate compositional reasoning descriptions implicit in visual features, and (iii) an Image-Text Dual Alignment (ITDA) module that aligns visual and textual features for fine-grained relational reasoning. This enables CORTEX to reason over visual and textual features and capture changes that are otherwise ambiguous in visual features alone.	Accep... Accepted at AAAI 2026	None
SAMChat: Introducing Chain of Thought Reasoning and GRPO to a Multimodal Small Language Model for Small Scale Remote Sensing	2025-11-27	Show Remarkable capabilities in understanding and generating text-image content have been demonstrated by recent advancements in multimodal large language models (MLLMs). However, their effectiveness in specialized domains-particularly those requiring resource-efficient and domain-specific adaptations-has remained limited. In this work, a lightweight multimodal language model termed SAMChat is introduced, specifically adapted to analyze remote sensing imagery in secluded areas, including challenging missile launch sites. A new dataset, SAMData, was compiled by verifying hundreds of aerial images through expert review, and subtle military installations were highlighted via detailed captions. Supervised fine-tuning on a 2B parameter open-source MLLM with chain-of-thought (CoT) reasoning annotations was performed, enabling more accurate and interpretable explanations. Additionally, Group Relative Policy Optimization (GRPO) was leveraged to enhance the model's ability to detect critical domain-specific cues-such as defensive layouts and key military structures-while minimizing false positives on civilian scenes. Through empirical evaluations, it has been shown that SAMChat significantly outperforms both larger, general-purpose multimodal models and existing remote sensing adapted approaches on open-ended captioning and classification metrics. Over 80% recall and 98% precision were achieved on the newly proposed SAMData benchmark, underscoring the potency of targeted fine-tuning and reinforcement learning in specialized real-world applications.	Accep... Accepted to Journal of Selected Topics in Applied Earth Observations and Remote Sensing (JSTARS) Special Issue on Foundation and Large Vision Models for Remote Sensing. Code and dataset are available at https://github.com/aybora/SAMChat	Code Link
CLIP-like Model as a Foundational Density Ratio Estimator	2025-11-27	Show Density ratio estimation is a core concept in statistical machine learning because it provides a unified mechanism for tasks such as importance weighting, divergence estimation, and likelihood-free inference, but its potential in vision and language models has not been fully explored. Modern vision-language encoders such as CLIP and SigLIP are trained with contrastive objectives that implicitly optimize log density ratios between joint and marginal image-text distributions, which implicitly learn similarity scores proportional to log density ratios. However, prior work has largely focused on their embedding utility, and the density-ratio structure induced by contrastive learning has not been systematically examined or exploited in multimodal applications. To address this gap, we reinterpret CLIP-style models as pretrained and general-purpose density ratio estimators and show that this perspective enables new algorithmic capabilities. We present a unified explanation of how contrastive objectives estimate density ratios and propose two practical applications: Importance Weight Learning and KL divergence estimation. Our Importance Weight Learning method requires only a single additional prompt and improves F1 scores by up to 7 points. We further show that CLIP-based density ratios support estimation of KL divergences that quantify how conditioning on an image or text alters the distribution of the other modality. Through qualitative examples and an N-gram analysis of captions, we find that these divergences capture semantic diversity and mode structure in multimodal data. Leveraging this property, we introduce a simple KL-guided data curation method that achieves performance competitive with LAION2B filtering.		None
Bridging the Modality Gap by Similarity Standardization with Pseudo-Positive Samples	2025-11-27	Show Advances in vision-language models (VLMs) have enabled effective cross-modality retrieval. However, when both text and images exist in the database, similarity scores would differ in scale by modality. This phenomenon, known as the modality gap, hinders accurate retrieval. Most existing studies address this issue with manually labeled data, e.g., by fine-tuning VLMs on them. In this work, we propose a similarity standardization approach with pseudo data construction. We first compute the mean and variance of the similarity scores between each query and its paired data in text or image modality. Using these modality-specific statistics, we standardize all similarity scores to compare on a common scale across modalities. These statistics are calculated from pseudo pairs, which are constructed by retrieving the text and image candidates with the highest cosine similarity to each query. We evaluate our method across seven VLMs using two multi-modal QA benchmarks (MMQA and WebQA), where each question requires retrieving either text or image data. Our experimental results show that our method significantly improves retrieval performance, achieving average Recall@20 gains of 64% on MMQA and 28% on WebQA when the query and the target data belong to different modalities. Compared to E5-V, which addresses the modality gap through image captioning, we confirm that our method more effectively bridges the modality gap.	Accep... Accepted to PACLIC2025	None
PROMPTMINER: Black-Box Prompt Stealing against Text-to-Image Generative Models via Reinforcement Learning and Fuzz Optimization	2025-11-27	Show Text-to-image (T2I) generative models such as Stable Diffusion and FLUX can synthesize realistic, high-quality images directly from textual prompts. The resulting image quality depends critically on well-crafted prompts that specify both subjects and stylistic modifiers, which have become valuable digital assets. However, the rising value and ubiquity of high-quality prompts expose them to security and intellectual-property risks. One key threat is the prompt stealing attack, i.e., the task of recovering the textual prompt that generated a given image. Prompt stealing enables unauthorized extraction and reuse of carefully engineered prompts, yet it can also support beneficial applications such as data attribution, model provenance analysis, and watermarking validation. Existing approaches often assume white-box gradient access, require large-scale labeled datasets for supervised training, or rely solely on captioning without explicit optimization, limiting their practicality and adaptability. To address these challenges, we propose PROMPTMINER, a black-box prompt stealing framework that decouples the task into two phases: (1) a reinforcement learning-based optimization phase to reconstruct the primary subject, and (2) a fuzzing-driven search phase to recover stylistic modifiers. Experiments across multiple datasets and diffusion backbones demonstrate that PROMPTMINER achieves superior results, with CLIP similarity up to 0.958 and textual alignment with SBERT up to 0.751, surpassing all baselines. Even when applied to in-the-wild images with unknown generators, it outperforms the strongest baseline by 7.5 percent in CLIP similarity, demonstrating better generalization. Finally, PROMPTMINER maintains strong performance under defensive perturbations, highlighting remarkable robustness. Code: https://github.com/aaFrostnova/PromptMiner		Code Link
WalkCLIP: Multimodal Learning for Urban Walkability Prediction	2025-11-26	Show Urban walkability is a cornerstone of public health, sustainability, and quality of life. Traditional walkability assessments rely on surveys and field audits, which are costly and difficult to scale. Recent studies have used satellite imagery, street view imagery, or population indicators to estimate walkability, but these single-source approaches capture only one dimension of the walking environment. Satellite data describe the built environment from above, but overlook the pedestrian perspective. Street view imagery captures conditions at the ground level, but lacks broader spatial context. Population dynamics reveal patterns of human activity but not the visual form of the environment. We introduce WalkCLIP, a multimodal framework that integrates these complementary viewpoints to predict urban walkability. WalkCLIP learns walkability-aware vision-language representations from GPT-4o generated image captions, refines these representations with a spatial aggregation module that incorporates neighborhood context, and fuses the resulting features with representations from a population dynamics foundation model. Evaluated at 4,660 locations throughout Minneapolis-Saint Paul, WalkCLIP outperforms unimodal and multimodal baselines in both predictive accuracy and spatial alignment. These results show that the integration of visual and behavioral signals yields reliable predictions of the walking environment.		None
SAM Guided Semantic and Motion Changed Region Mining for Remote Sensing Change Captioning	2025-11-26	Show Remote sensing change captioning is an emerging and popular research task that aims to describe, in natural language, the content of interest that has changed between two remote sensing images captured at different times. Existing methods typically employ CNNs/Transformers to extract visual representations from the given images or incorporate auxiliary tasks to enhance the final results, with weak region awareness and limited temporal alignment. To address these issues, this paper explores the use of the SAM (Segment Anything Model) foundation model to extract region-level representations and inject region-of-interest knowledge into the captioning framework. Specifically, we employ a CNN/Transformer model to extract global-level vision features, leverage the SAM foundation model to delineate semantic- and motion-level change regions, and utilize a specially constructed knowledge graph to provide information about objects of interest. These heterogeneous sources of information are then fused via cross-attention, and a Transformer decoder is used to generate the final natural language description of the observed changes. Extensive experimental results demonstrate that our method achieves state-of-the-art performance across multiple widely used benchmark datasets. The source code of this paper will be released on https://github.com/Event-AHU/SAM_ChangeCaptioning		Code Link
BanglaMM-Disaster: A Multimodal Transformer-Based Deep Learning Framework for Multiclass Disaster Classification in Bangla	2025-11-26	Show Natural disasters remain a major challenge for Bangladesh, so real-time monitoring and quick response systems are essential. In this study, we present BanglaMM-Disaster, an end-to-end deep learning-based multimodal framework for disaster classification in Bangla, using both textual and visual data from social media. We constructed a new dataset of 5,037 Bangla social media posts, each consisting of a caption and a corresponding image, annotated into one of nine disaster-related categories. The proposed model integrates transformer-based text encoders, including BanglaBERT, mBERT, and XLM-RoBERTa, with CNN backbones such as ResNet50, DenseNet169, and MobileNetV2, to process the two modalities. Using early fusion, the best model achieves 83.76% accuracy. This surpasses the best text-only baseline by 3.84% and the image-only baseline by 16.91%. Our analysis also shows reduced misclassification across all classes, with noticeable improvements for ambiguous examples. This work fills a key gap in Bangla multimodal disaster analysis and demonstrates the benefits of combining multiple data types for real-time disaster response in low-resource settings.	Prese... Presented at the 2025 IEEE International Conference on Signal Processing, Information, Communication and Systems (SPICSCON), November 21-22, 2025, University of Rajshahi, Bangladesh. 6 pages, 9 disaster classes, multimodal dataset with 5,037 samples	None
SARVLM: A Vision Language Foundation Model for Semantic Understanding and Target Recognition in SAR Imagery	2025-11-26	Show Synthetic Aperture Radar (SAR) is a crucial imaging modality thanks to its all-weather capability. Although recent advances in self-supervised learning and masked image modeling (MIM) have enabled SAR foundation models, these methods largely emphasize low-level visual features and often overlook multimodal alignment and zero-shot target recognition in SAR imagery. To address this, we construct SARVLM-1M, a large-scale vision-language dataset with over one million image-text pairs aggregated from existing datasets. We further propose a domain transfer training strategy to mitigate the large gap between natural and SAR imagery. Building on this, we develop SARVLM, the first vision language foundation model (VLM) tailored to SAR, comprising SARCLIP and SARCap. SARVLM is trained with a vision-language contrastive objective under the proposed domain transfer strategy, bridging SAR imagery and textual descriptions. Extensive experiments on image text retrieval, zero-shot classification, semantic localization, and imagery captioning demonstrate that SARVLM delivers superior feature extraction and interpretation, outperforming state-of-the-art VLMs and advancing SAR semantic understanding. Code and datasets will be released soon.	11 pages, 9 figures	None
CAPability: A Comprehensive Visual Caption Benchmark for Evaluating Both Correctness and Thoroughness	2025-11-26	Show Visual captioning benchmarks have become outdated with the emergence of modern multimodal large language models (MLLMs), as the brief ground-truth sentences and traditional metrics fail to assess detailed captions effectively. While recent benchmarks attempt to address this by focusing on keyword extraction or object-centric evaluation, they remain limited to vague-view or object-view analyses and incomplete visual element coverage. In this paper, we introduce CAPability, a comprehensive multi-view benchmark for evaluating visual captioning across 12 dimensions spanning six critical views. We curate nearly 11K human-annotated images and videos with visual element annotations to evaluate the generated captions. CAPability stably assesses both the correctness and thoroughness of captions with \textit{precision} and \textit{hit} metrics. By converting annotations to QA pairs, we further introduce a heuristic metric, \textit{know but cannot tell} ($K\bar{T}$), indicating a significant performance gap between QA and caption capabilities. Our work provides a holistic analysis of MLLMs' captioning abilities, as we identify their strengths and weaknesses across various dimensions, guiding future research to enhance specific aspects of their capabilities.	Accep... Accepted to NeurIPS 2025	None
CaptionQA: Is Your Caption as Useful as the Image Itself?	2025-11-26	Show Image captions serve as efficient surrogates for visual content in multimodal systems such as retrieval, recommendation, and multi-step agentic inference pipelines. Yet current evaluation practices miss a fundamental question: Can captions stand-in for images in real downstream tasks? We propose a utility-based benchmark, CaptionQA, to evaluate model-generated captions, where caption quality is measured by how well it supports downstream tasks. CaptionQA is an extensible domain-dependent benchmark covering 4 domains--Natural, Document, E-commerce, and Embodied AI--each with fine-grained taxonomies (25 top-level and 69 subcategories) that identify useful information for domain-specific tasks. CaptionQA builds 33,027 densely annotated multiple-choice questions (50.3 per image on average) that explicitly require visual information to answer, providing a comprehensive probe of caption utility. In our evaluation protocol, an LLM answers these questions using captions alone, directly measuring whether captions preserve image-level utility and are utilizable by a downstream LLM. Evaluating state-of-the-art MLLMs reveals substantial gaps between the image and its caption utility. Notably, models nearly identical on traditional image-QA benchmarks lower by up to 32% in caption utility. We release CaptionQA along with an open-source pipeline for extension to new domains. The code is available at https://github.com/bronyayang/CaptionQA.		Code Link
Knowledge Completes the Vision: A Multimodal Entity-aware Retrieval-Augmented Generation Framework for News Image Captioning	2025-11-26	Show News image captioning aims to produce journalistically informative descriptions by combining visual content with contextual cues from associated articles. Despite recent advances, existing methods struggle with three key challenges: (1) incomplete information coverage, (2) weak cross-modal alignment, and (3) suboptimal visual-entity grounding. To address these issues, we introduce MERGE, the first Multimodal Entity-aware Retrieval-augmented GEneration framework for news image captioning. MERGE constructs an entity-centric multimodal knowledge base (EMKB) that integrates textual, visual, and structured knowledge, enabling enriched background retrieval. It improves cross-modal alignment through a multistage hypothesis-caption strategy and enhances visual-entity matching via dynamic retrieval guided by image content. Extensive experiments on GoodNews and NYTimes800k show that MERGE significantly outperforms state-of-the-art baselines, with CIDEr gains of +6.84 and +1.16 in caption quality, and F1-score improvements of +4.14 and +2.64 in named entity recognition. Notably, MERGE also generalizes well to the unseen Visual News dataset, achieving +20.17 in CIDEr and +6.22 in F1-score, demonstrating strong robustness and domain adaptability.	Accep... Accepted to AAAI 2026	None
Beyond Generation: Multi-Hop Reasoning for Factual Accuracy in Vision-Language Models	2025-11-25	Show Visual Language Models (VLMs) are powerful generative tools but often produce factually inaccurate outputs due to a lack of robust reasoning capabilities. While extensive research has been conducted on integrating external knowledge for reasoning in large language models (LLMs), such efforts remain underexplored in VLMs, where the challenge is compounded by the need to bridge multiple modalities seamlessly. This work introduces a framework for knowledge-guided reasoning in VLMs, leveraging structured knowledge graphs for multi-hop verification using image-captioning task to illustrate our framework. Our approach enables systematic reasoning across multiple steps, including visual entity recognition, knowledge graph traversal, and fact-based caption refinement. We evaluate the framework using hierarchical, triple-based and bullet-point based knowledge representations, analyzing their effectiveness in factual accuracy and logical inference. Empirical results show that our approach improves factual accuracy by approximately 31% on preliminary experiments on a curated dataset of mixtures from Google Landmarks v2, Conceptual captions and Coco captions revealing key insights into reasoning patterns and failure modes. This work demonstrates the potential of integrating external knowledge for advancing reasoning in VLMs, paving the way for more reliable and knowledgable multimodal systems.	Accep... Accepted as poster at NewInML Workshop ICML, 2025	None
AlignBench: Benchmarking Fine-Grained Image-Text Alignment with Synthetic Image-Caption Pairs	2025-11-25	Show Assessing image-text alignment models such as CLIP is crucial for bridging visual and linguistic representations. Yet existing benchmarks rely on rule-based perturbations or short captions, limiting their ability to measure fine-grained alignment. We introduce AlignBench, a benchmark that provides a new indicator of image-text alignment by evaluating detailed image-caption pairs generated by diverse image-to-text and text-to-image models. Each sentence is annotated for correctness, enabling direct assessment of VLMs as alignment evaluators. Benchmarking a wide range of decoder-based VLMs reveals three key findings: (i) CLIP-based models, even those tailored for compositional reasoning, remain nearly blind; (ii) detectors systematically over-score early sentences; and (iii) they show strong self-preference, favoring their own outputs and harming detection performance. Our project page will be available at https://dahlian00.github.io/AlignBench/.	Proje... Project Page: https://dahlian00.github.io/AlignBench/	Code Link
Panoptic Captioning: An Equivalence Bridge for Image and Text	2025-11-25	Show This work introduces panoptic captioning, a novel task striving to seek the minimum text equivalent of images, which has broad potential applications. We take the first step towards panoptic captioning by formulating it as a task of generating a comprehensive textual description for an image, which encapsulates all entities, their respective locations and attributes, relationships among entities, as well as global image state. Through an extensive evaluation, our work reveals that state-of-the-art Multi-modal Large Language Models (MLLMs) have limited performance in solving panoptic captioning. To address this, we propose an effective data engine named PancapEngine to produce high-quality data and a novel method named PancapChain to improve panoptic captioning. Specifically, our PancapEngine first detects diverse categories of entities in images by an elaborate detection suite, and then generates required panoptic captions using entity-aware prompts. Additionally, our PancapChain explicitly decouples the challenging panoptic captioning task into multiple stages and generates panoptic captions step by step. More importantly, we contribute a comprehensive metric named PancapScore and a human-curated test set for reliable model evaluation. Experiments show that our PancapChain-13B model can beat state-of-the-art open-source MLLMs like InternVL-2.5-78B and even surpass proprietary models like GPT-4o and Gemini-2.0-Pro, demonstrating the effectiveness of our data engine and method. Project page: https://visual-ai.github.io/pancap/	NeurI... NeurIPS 2025; Project page: https://visual-ai.github.io/pancap/	Code Link
Tell Model Where to Look: Mitigating Hallucinations in MLLMs by Vision-Guided Attention	2025-11-25	Show Visual attention serves as the primary mechanism through which MLLMs interpret visual information; however, its limited localization capability often leads to hallucinations. We observe that although MLLMs can accurately extract visual semantics from visual tokens, they fail to fully leverage this advantage during subsequent inference. To address this limitation, we propose Vision-Guided Attention (VGA), a training-free method that first constructs precise visual grounding by exploiting the semantic content of visual tokens, and then uses this grounding to guide the model's focus toward relevant visual regions. In image captioning, VGA further refines this guidance dynamically during generation by suppressing regions that have already been described. In VGA, each token undergoes only a single forward pass, introducing a negligible latency overhead of just 4.36%. In addition, VGA is fully compatible with efficient attention implementations such as FlashAttention. Extensive experiments across diverse MLLMs and multiple hallucination benchmarks demonstrate that VGA achieves state-of-the-art dehallucination performance. Further analysis confirms that explicit visual guidance plays a crucial role in enhancing the visual understanding capabilities of MLLMs.	Under Review	None
Cross-Layer Vision Smoothing: Enhancing Visual Understanding via Sustained Focus on Key Objects in Large Vision-Language Models	2025-11-25	Show Large Vision-Language Models (LVLMs) can accurately locate key objects in images, yet their attention to these objects tends to be very brief. Motivated by the hypothesis that sustained focus on key objects can improve LVLMs' visual capabilities, we propose Cross-Layer Vision Smoothing (CLVS). The core idea of CLVS is to incorporate a vision memory that smooths the attention distribution across layers. Specifically, we initialize this vision memory with position-unbiased visual attention in the first layer. In subsequent layers, the model's visual attention jointly considers the vision memory from previous layers, while the memory is updated iteratively, thereby maintaining smooth attention on key objects. Given that visual understanding primarily occurs in the early and middle layers of the model, we use uncertainty as an indicator of completed visual understanding and terminate the smoothing process accordingly. Experiments on four benchmarks across three LVLMs confirm the effectiveness and generalizability of our method. CLVS achieves state-of-the-art overall performance across a variety of visual understanding tasks and attains comparable results to the leading approaches on image captioning benchmarks.	Under Review	None
Are Neuro-Inspired Multi-Modal Vision-Language Models Resilient to Membership Inference Privacy Leakage?	2025-11-24	Show In the age of agentic AI, the growing deployment of multi-modal models (MMs) has introduced new attack vectors that can leak sensitive training data in MMs, causing privacy leakage. This paper investigates a black-box privacy attack, i.e., membership inference attack (MIA) on multi-modal vision-language models (VLMs). State-of-the-art research analyzes privacy attacks primarily to unimodal AI-ML systems, while recent studies indicate MMs can also be vulnerable to privacy attacks. While researchers have demonstrated that biologically inspired neural network representations can improve unimodal model resilience against adversarial attacks, it remains unexplored whether neuro-inspired MMs are resilient against privacy attacks. In this work, we introduce a systematic neuroscience-inspired topological regularization (tau) framework to analyze MM VLMs resilience against image-text-based inference privacy attacks. We examine this phenomenon using three VLMs: BLIP, PaliGemma 2, and ViT-GPT2, across three benchmark datasets: COCO, CC3M, and NoCaps. Our experiments compare the resilience of baseline and neuro VLMs (with topological regularization), where the tau > 0 configuration defines the NEURO variant of VLM. Our results on the BLIP model using the COCO dataset illustrate that MIA attack success in NEURO VLMs drops by 24% mean ROC-AUC, while achieving similar model utility (similarities between generated and reference captions) in terms of MPNet and ROUGE-2 metrics. This shows neuro VLMs are comparatively more resilient against privacy attacks, while not significantly compromising model utility. Our extensive evaluation with PaliGemma 2 and ViT-GPT2 models, on two additional datasets: CC3M and NoCaps, further validates the consistency of the findings. This work contributes to the growing understanding of privacy risks in MMs and provides evidence on neuro VLMs privacy threat resilience.		None
Vision Language Models Can Parse Floor Plan Maps	2025-11-24	Show Vision language models (VLMs) can simultaneously reason about images and texts to tackle many tasks, from visual question answering to image captioning. This paper focuses on map parsing, a novel task that is unexplored within the VLM context and particularly useful to mobile robots. Map parsing requires understanding not only the labels but also the geometric configurations of a map, i.e., what areas are like and how they are connected. To evaluate the performance of VLMs on map parsing, we prompt VLMs with floor plan maps to generate task plans for complex indoor navigation. Our results demonstrate the remarkable capability of VLMs in map parsing, with a success rate of 0.96 in tasks requiring a sequence of nine navigation actions, e.g., approaching and going through doors. Other than intuitive observations, e.g., VLMs do better in smaller maps and simpler navigation tasks, there was a very interesting observation that its performance drops in large open areas. We provide practical suggestions to address such challenges as validated by our experimental results. Webpage: https://sites.google.com/view/vlm-floorplan/		None
Connecting the Dots: Training-Free Visual Grounding via Agentic Reasoning	2025-11-24	Show Visual grounding, the task of linking textual queries to specific regions within images, plays a pivotal role in vision-language integration. Existing methods typically rely on extensive task-specific annotations and fine-tuning, limiting their ability to generalize effectively to novel or out-of-distribution scenarios. To address these limitations, we introduce GroundingAgent, a novel agentic visual grounding framework that operates without any task-specific fine-tuning. GroundingAgent employs a structured, iterative reasoning mechanism that integrates pretrained open-vocabulary object detectors, multimodal large language models (MLLMs), and large language models (LLMs) to progressively refine candidate regions through joint semantic and spatial analyses. Remarkably, GroundingAgent achieves an average zero-shot grounding accuracy of 65.1 % on widely-used benchmarks (RefCOCO, RefCOCO+, RefCOCOg), entirely without fine-tuning. Furthermore, by substituting MLLM-generated captions with the original query texts, the accuracy at the selection stage alone reaches approximately 90 %, closely matching supervised performance and underscoring the critical role of LLM reasoning capabilities. GroundingAgent also offers strong interpretability, transparently illustrating each reasoning step and providing clear insights into its decision-making process.	AAAI 2025	None
Can Modern Vision Models Understand the Difference Between an Object and a Look-alike?	2025-11-24	Show Recent advances in computer vision have yielded models with strong performance on recognition benchmarks; however, significant gaps remain in comparison to human perception. One subtle ability is to judge whether an image looks like a given object without being an instance of that object. We study whether vision-language models such as CLIP capture this distinction. We curated a dataset named RoLA (Real or Lookalike) of real and lookalike exemplars (e.g., toys, statues, drawings, pareidolia) across multiple categories, and first evaluate a prompt-based baseline with paired "real"/"lookalike" prompts. We then estimate a direction in CLIP's embedding space that moves representations between real and lookalike. Applying this direction to image and text embeddings improves discrimination in cross-modal retrieval on Conceptual12M, and also enhances captions produced by a CLIP prefix captioner.		None
CLASH: A Benchmark for Cross-Modal Contradiction Detection	2025-11-24	Show Contradictory multimodal inputs are common in real-world settings, yet existing benchmarks typically assume input consistency and fail to evaluate cross-modal contradiction detection - a fundamental capability for preventing hallucinations and ensuring reliability. We introduce CLASH, a novel benchmark for multimodal contradiction detection, featuring COCO images paired with contradictory captions containing controlled object-level or attribute-level contradictions. The samples include targeted questions evaluated in both multiple-choice and open-ended formats. The benchmark provides an extensive fine-tuning set filtered through automated quality checks, alongside a smaller human-verified diagnostic set. Our analysis of state-of-the-art models reveals substantial limitations in recognizing cross-modal conflicts, exposing systematic modality biases and category-specific weaknesses. Furthermore, we empirically demonstrate that targeted fine-tuning on CLASH substantially enhances conflict detection capabilities.	First... First two authors contributed equally	None
BackdoorVLM: A Benchmark for Backdoor Attacks on Vision-Language Models	2025-11-24	Show Backdoor attacks undermine the reliability and trustworthiness of machine learning systems by injecting hidden behaviors that can be maliciously activated at inference time. While such threats have been extensively studied in unimodal settings, their impact on multimodal foundation models, particularly vision-language models (VLMs), remains largely underexplored. In this work, we introduce \textbf{BackdoorVLM}, the first comprehensive benchmark for systematically evaluating backdoor attacks on VLMs across a broad range of settings. It adopts a unified perspective that injects and analyzes backdoors across core vision-language tasks, including image captioning and visual question answering. BackdoorVLM organizes multimodal backdoor threats into 5 representative categories: targeted refusal, malicious injection, jailbreak, concept substitution, and perceptual hijack. Each category captures a distinct pathway through which an adversary can manipulate a model's behavior. We evaluate these threats using 12 representative attack methods spanning text, image, and bimodal triggers, tested on 2 open-source VLMs and 3 multimodal datasets. Our analysis reveals that VLMs exhibit strong sensitivity to textual instructions, and in bimodal backdoors the text trigger typically overwhelms the image trigger when forming the backdoor mapping. Notably, backdoors involving the textual modality remain highly potent, with poisoning rates as low as 1% yielding over 90% success across most tasks. These findings highlight significant, previously underexplored vulnerabilities in current VLMs. We hope that BackdoorVLM can serve as a useful benchmark for analyzing and mitigating multimodal backdoor threats. Code is available at: https://github.com/bin015/BackdoorVLM .		Code Link
ProxT2I: Efficient Reward-Guided Text-to-Image Generation via Proximal Diffusion	2025-11-24	Show Diffusion models have emerged as a dominant paradigm for generative modeling across a wide range of domains, including prompt-conditional generation. The vast majority of samplers, however, rely on forward discretization of the reverse diffusion process and use score functions that are learned from data. Such forward and explicit discretizations can be slow and unstable, requiring a large number of sampling steps to produce good-quality samples. In this work we develop a text-to-image (T2I) diffusion model based on backward discretizations, dubbed ProxT2I, relying on learned and conditional proximal operators instead of score functions. We further leverage recent advances in reinforcement learning and policy optimization to optimize our samplers for task-specific rewards. Additionally, we develop a new large-scale and open-source dataset comprising 15 million high-quality human images with fine-grained captions, called LAION-Face-T2I-15M, for training and evaluation. Our approach consistently enhances sampling efficiency and human-preference alignment compared to score-based baselines, and achieves results on par with existing state-of-the-art and open-source text-to-image models while requiring lower compute and smaller model size, offering a lightweight yet performant solution for human text-to-image generation.		None
Find Them All: Unveiling MLLMs for Versatile Person Re-identification	2025-11-23	Show Person re-identification (ReID) aims to retrieve images of a target person from the gallery set, with wide applications in medical rehabilitation and public security. However, traditional person ReID models are typically uni-modal, resulting in limited generalizability across heterogeneous data modalities. Recently, the emergence of multi-modal large language models (MLLMs) has shown a promising avenue for addressing this issue. Despite this potential, existing methods merely regard MLLMs as feature extractors or caption generators, leaving their capabilities in person ReID tasks largely unexplored. To bridge this gap, we introduce a novel benchmark for \underline{\textbf{V}}ersatile \underline{\textbf{P}}erson \underline{\textbf{Re}}-\underline{\textbf{ID}}entification, termed VP-ReID. The benchmark includes 257,310 multi-modal queries and gallery images, covering ten diverse person ReID tasks. In addition, we propose two task-oriented evaluation schemes for MLLM-based person ReID. Extensive experiments demonstrate the impressive versatility, effectiveness, and interpretability of MLLMs in various person ReID tasks. Nevertheless, they also have limitations in handling a few modalities, particularly thermal and infrared data. We hope that VP-ReID can facilitate the community in developing more robust and generalizable cross-modal foundation models for person ReID.		None
Can a Second-View Image Be a Language? Geometric and Semantic Cross-Modal Reasoning for X-ray Prohibited Item Detection	2025-11-23	Show Automatic X-ray prohibited items detection is vital for security inspection and has been widely studied. Traditional methods rely on visual modality, often struggling with complex threats. While recent studies incorporate language to guide single-view images, human inspectors typically use dual-view images in practice. This raises the question: can the second view provide constraints similar to a language modality? In this work, we introduce DualXrayBench, the first comprehensive benchmark for X-ray inspection that includes multiple views and modalities. It supports eight tasks designed to test cross-view reasoning. In DualXrayBench, we introduce a caption corpus consisting of 45,613 dual-view image pairs across 12 categories with corresponding captions. Building upon these data, we propose the Geometric (cross-view)-Semantic (cross-modality) Reasoner (GSR), a multimodal model that jointly learns correspondences between cross-view geometry and cross-modal semantics, treating the second-view images as a "language-like modality". To enable this, we construct the GSXray dataset, with structured Chain-of-Thought sequences: , , . Comprehensive evaluations on DualXrayBench demonstrate that GSR achieves significant improvements across all X-ray tasks, offering a new perspective for real-world X-ray inspection.	10 pages, 4 figures	None
UltraFlux: Data-Model Co-Design for High-quality Native 4K Text-to-Image Generation across Diverse Aspect Ratios	2025-11-22	Show Diffusion transformers have recently delivered strong text-to-image generation around 1K resolution, but we show that extending them to native 4K across diverse aspect ratios exposes a tightly coupled failure mode spanning positional encoding, VAE compression, and optimization. Tackling any of these factors in isolation leaves substantial quality on the table. We therefore take a data-model co-design view and introduce UltraFlux, a Flux-based DiT trained natively at 4K on MultiAspect-4K-1M, a 1M-image 4K corpus with controlled multi-AR coverage, bilingual captions, and rich VLM/IQA metadata for resolution- and AR-aware sampling. On the model side, UltraFlux couples (i) Resonance 2D RoPE with YaRN for training-window-, frequency-, and AR-aware positional encoding at 4K; (ii) a simple, non-adversarial VAE post-training scheme that improves 4K reconstruction fidelity; (iii) an SNR-Aware Huber Wavelet objective that rebalances gradients across timesteps and frequency bands; and (iv) a Stage-wise Aesthetic Curriculum Learning strategy that concentrates high-aesthetic supervision on high-noise steps governed by the model prior. Together, these components yield a stable, detail-preserving 4K DiT that generalizes across wide, square, and tall ARs. On the Aesthetic-Eval at 4096 benchmark and multi-AR 4K settings, UltraFlux consistently outperforms strong open-source baselines across fidelity, aesthetic, and alignment metrics, and-with a LLM prompt refiner-matches or surpasses the proprietary Seedream 4.0.	Proje... Project Page: https://w2genai-lab.github.io/UltraFlux/	Code Link
HUMORCHAIN: Theory-Guided Multi-Stage Reasoning for Interpretable Multimodal Humor Generation	2025-11-21	Show Humor, as both a creative human activity and a social binding mechanism, has long posed a major challenge for AI generation. Although producing humor requires complex cognitive reasoning and social understanding, theories of humor suggest that it follows learnable patterns and structures, making it theoretically possible for generative models to acquire them implicitly. In recent years, multimodal humor has become a prevalent form of online communication, especially among Gen Z, highlighting the need for AI systems capable of integrating visual understanding with humorous language generation. However, existing data-driven approaches lack explicit modeling or theoretical grounding of humor, often producing literal descriptions that fail to capture its underlying cognitive mechanisms, resulting in the generated image descriptions that are fluent but lack genuine humor or cognitive depth. To address this limitation, we propose HUMORCHAIN (HUmor-guided Multi-step Orchestrated Reasoning Chain for Image Captioning), a theory-guided multi-stage reasoning framework. It integrates visual semantic parsing, humor- and psychology-based reasoning, and a fine-tuned discriminator for humor evaluation, forming an interpretable and controllable cognitive reasoning chain. To the best of our knowledge, this is the first work to explicitly embed cognitive structures from humor theories into multimodal humor generation, enabling a structured reasoning process from visual understanding to humor creation. Experiments on Meme-Image-No-Text, Oogiri-GO, and OxfordTVG-HIC datasets show that HUMORCHAIN outperforms state-of-the-art baselines in human humor preference, Elo/BT scores, and semantic diversity, demonstrating that theory-driven structured reasoning enables large language models to generate humor aligned with human perception.		None
Aligning Vision to Language: Annotation-Free Multimodal Knowledge Graph Construction for Enhanced LLMs Reasoning	2025-11-21	Show Multimodal reasoning in Large Language Models (LLMs) struggles with incomplete knowledge and hallucination artifacts, challenges that textual Knowledge Graphs (KGs) only partially mitigate due to their modality isolation. While Multimodal Knowledge Graphs (MMKGs) promise enhanced cross-modal understanding, their practical construction is impeded by semantic narrowness of manual text annotations and inherent noise in visual-semantic entity linkages. In this paper, we propose Vision-align-to-Language integrated Knowledge Graph (VaLiK), a novel approach for constructing MMKGs that enhances LLMs reasoning through cross-modal information supplementation. Specifically, we cascade pre-trained Vision-Language Models (VLMs) to align image features with text, transforming them into descriptions that encapsulate image-specific information. Furthermore, we developed a cross-modal similarity verification mechanism to quantify semantic consistency, effectively filtering out noise introduced during feature alignment. Even without manually annotated image captions, the refined descriptions alone suffice to construct the MMKG. Compared to conventional MMKGs construction paradigms, our approach achieves substantial storage efficiency gains while maintaining direct entity-to-image linkage capability. Experimental results on multimodal reasoning tasks demonstrate that LLMs augmented with VaLiK outperform previous state-of-the-art models. Our code is published at https://github.com/Wings-Of-Disaster/VaLiK.	14 pa... 14 pages, 7 figures, 6 tables; Accepted by ICCV 2025	Code Link
OmniPT: Unleashing the Potential of Large Vision Language Models for Pedestrian Tracking and Understanding	2025-11-21	Show LVLMs have been shown to perform excellently in image-level tasks such as VQA and caption. However, in many instance-level tasks, such as visual grounding and object detection, LVLMs still show performance gaps compared to previous expert models. Meanwhile, although pedestrian tracking is a classical task, there have been a number of new topics in combining object tracking and natural language, such as Referring MOT, Cross-view Referring MOT, and Semantic MOT. These tasks emphasize that models should understand the tracked object at an advanced semantic level, which is exactly where LVLMs excel. In this paper, we propose a new unified Pedestrian Tracking framework, namely OmniPT, which can track, track based on reference and generate semantic understanding of tracked objects interactively. We address two issues: how to model the tracking task into a task that foundation models can perform, and how to make the model output formatted answers. To this end, we implement a training phase consisting of RL-Mid Training-SFT-RL. Based on the pre-trained weights of the LVLM, we first perform a simple RL phase to enable the model to output fixed and supervisable bounding box format. Subsequently, we conduct a mid-training phase using a large number of pedestrian-related datasets. Finally, we perform supervised fine-tuning on several pedestrian tracking datasets, and then carry out another RL phase to improve the model's tracking performance and enhance its ability to follow instructions. We conduct experiments on tracking benchmarks and the experimental results demonstrate that the proposed method can perform better than the previous methods.	AAAI 2026	None
UniModel: A Visual-Only Framework for Unified Multimodal Understanding and Generation	2025-11-21	Show We present UniModel, a unified generative model that jointly supports visual understanding and visual generation within a single pixel-to-pixel diffusion framework. Our goal is to achieve unification along three axes: the model, the tasks, and the representations. At the representation level, we eliminate modality discrepancies by mapping both text and images into a shared visual space: textual prompts are rendered as painted text images on a clean canvas, and all inputs and outputs are treated purely as RGB pixels. This yields a fully vision-native formulation of multimodal learning. At the task level, a broad range of vision-language problems are cast as pixel-to-pixel transformations in this visual space. For understanding tasks, the model takes an RGB image and produces a painted text image that visually encodes the semantic prediction. For generation tasks, painted text images serve as visual conditions that guide realistic and semantically aligned image synthesis. Captioning and text-to-image generation thus become different directions of the same underlying visual translation process. At the model level, we instantiate a single Unified Diffusion Transformer trained with rectified flow in pixel space. A shared backbone jointly learns bidirectional mappings between natural images and painted text images, with lightweight task embeddings to specify the desired direction. Experiments on text-to-image synthesis and image-to-text understanding demonstrate strong cross-modal alignment and emergent controllability such as cycle-consistent image-caption-image loops. Our initial exploration suggests that unifying model, tasks, and representations in a single visual space is a promising paradigm for general-purpose multimodal intelligence.		None
PEPPER: Perception-Guided Perturbation for Robust Backdoor Defense in Text-to-Image Diffusion Models	2025-11-20	Show Recent studies show that text to image (T2I) diffusion models are vulnerable to backdoor attacks, where a trigger in the input prompt can steer generation toward harmful or unintended content. To address this, we introduce PEPPER (PErcePtion Guided PERturbation), a backdoor defense that rewrites the caption into a semantically distant yet visually similar caption while adding unobstructive elements. With this rewriting strategy, PEPPER disrupt the trigger embedded in the input prompt, dilute the influence of trigger tokens and thereby achieve enhanced robustness. Experiments show that PEPPER is particularly effective against text encoder based attacks, substantially reducing attack success while preserving generation quality. Beyond this, PEPPER can be paired with any existing defenses yielding consistently stronger and generalizable robustness than any standalone method. Our code will be released on Github.		None
Contrastive vision-language learning with paraphrasing and negation	2025-11-20	Show Contrastive vision-language models continue to be the dominant approach for image and text retrieval. Contrastive Language-Image Pre-training (CLIP) trains two neural networks in contrastive manner to align their image and text embeddings in a shared latent space. Recent results evaluating CLIP on negated or paraphrased text have shown mixed performance because negation changes meaning radically with minimal lexical changes, while paraphrasing can create very different textual expressions with the same intended meaning. This poses a significant challenge for improving the evaluation results and alignment of vision-language models. To address this challenge, this paper evaluates the combination of paraphrasing and negation, proposes a new CLIP contrastive loss function accounting for both paraphrasing and negation, and applies LLM-generated training triples consisting of original, paraphrased and negated textual captions to CLIP-like training models. The approach, called SemCLIP, is shown to move paraphrased captions towards the original image embeddings while pushing negated captions further away in embedding space. Empirically, SemCLIP is shown to be capable of preserving CLIP's performance while increasing considerably the distances to negated captions. On the CC-Neg benchmark using an original over negation image-retrieval accuracy metric, SemCLIP improves accuracy from 68.1% to 78.1%. Although results are mixed when compared with CLIP on the Sugarcrepe++ benchmark, SemCLIP's performance is generally better than the models trained with negated captions. This robustness to negation extends to downstream zero-shot classification tasks where SemCLIP pre-trained on Sugarcrepe++ performs better than CLIP on all tested downstream tasks. These results indicate that SemCLIP can achieve significant robustness to semantic transformations.		None
MMVA: Multimodal Matching Based on Valence and Arousal across Images, Music, and Musical Captions	2025-11-20	Show We introduce Multimodal Matching based on Valence and Arousal (MMVA), a tri-modal encoder framework designed to capture emotional content across images, music, and musical captions. To support this framework, we expand the Image-Music-Emotion-Matching-Net (IMEMNet) dataset, creating IMEMNet-C which includes 24,756 images and 25,944 music clips with corresponding musical captions. We employ multimodal matching scores based on the continuous valence (emotional positivity) and arousal (emotional intensity) values. This continuous matching score allows for random sampling of image-music pairs during training by computing similarity scores from the valence-arousal values across different modalities. Consequently, the proposed approach achieves state-of-the-art performance in valence-arousal prediction tasks. Furthermore, the framework demonstrates its efficacy in various zeroshot tasks, highlighting the potential of valence and arousal predictions in downstream applications.	Paper... Paper accepted in Artificial Intelligence for Music workshop at AAAI 2025	None
Cross Modal Fine-Grained Alignment via Granularity-Aware and Region-Uncertain Modeling	2025-11-19	Show Fine-grained image-text alignment is a pivotal challenge in multimodal learning, underpinning key applications such as visual question answering, image captioning, and vision-language navigation. Unlike global alignment, fine-grained alignment requires precise correspondence between localized visual regions and textual tokens, often hindered by noisy attention mechanisms and oversimplified modeling of cross-modal relationships. In this work, we identify two fundamental limitations of existing approaches: the lack of robust intra-modal mechanisms to assess the significance of visual and textual tokens, leading to poor generalization in complex scenes; and the absence of fine-grained uncertainty modeling, which fails to capture the one-to-many and many-to-one nature of region-word correspondences. To address these issues, we propose a unified approach that incorporates significance-aware and granularity-aware modeling and region-level uncertainty modeling. Our method leverages modality-specific biases to identify salient features without relying on brittle cross-modal attention, and represents region features as a mixture of Gaussian distributions to capture fine-grained uncertainty. Extensive experiments on Flickr30K and MS-COCO demonstrate that our approach achieves state-of-the-art performance across various backbone architectures, significantly enhancing the robustness and interpretability of fine-grained image-text alignment.	10 pa... 10 pages, 6 figures, accepted by AAAI 2026	None
Continual Learning for Image Captioning through Improved Image-Text Alignment	2025-11-18	Show Generating accurate and coherent image captions in a continual learning setting remains a major challenge due to catastrophic forgetting and the difficulty of aligning evolving visual concepts with language over time. In this work, we propose a novel multi-loss framework for continual image captioning that integrates semantic guidance through prompt-based continual learning and contrastive alignment. Built upon a pretrained ViT-GPT-2 backbone, our approach combines standard cross-entropy loss with three additional components: (1) a prompt-based cosine similarity loss that aligns image embeddings with synthetically constructed prompts encoding objects, attributes, and actions; (2) a CLIP-style loss that promotes alignment between image embeddings and target caption embedding; and (3) a language-guided contrastive loss that employs a triplet loss to enhance class-level discriminability between tasks. Notably, our approach introduces no additional overhead at inference time and requires no prompts during caption generation. We find that this approach mitigates catastrophic forgetting, while achieving better semantic caption alignment compared to state-of-the-art methods. The code can be found via the following link: https://github.com/Gepardius/Taetz_Bordelius_Continual_ImageCaptioning.	11 pages, 3 figures	Code Link
FarSLIP: Discovering Effective CLIP Adaptation for Fine-Grained Remote Sensing Understanding	2025-11-18	Show As CLIP's global alignment limits its ability to capture fine-grained details, recent efforts have focused on enhancing its region-text alignment. However, current remote sensing (RS)-specific CLIP variants still inherit this limited spatial awareness. We identify two key limitations behind this: (1) current RS image-text datasets generate global captions from object-level labels, leaving the original object-level supervision underutilized; (2) despite the success of region-text alignment methods in general domain, their direct application to RS data often leads to performance degradation. To address these, we construct the first multi-granularity RS image-text dataset, MGRS-200k, featuring rich object-level textual supervision for RS region-category alignment. We further investigate existing fine-grained CLIP tuning strategies and find that current explicit region-text alignment methods, whether in a direct or indirect way, underperform due to severe degradation of CLIP's semantic coherence. Building on these, we propose FarSLIP, a Fine-grained Aligned RS Language-Image Pretraining framework. Rather than the commonly used patch-to-CLS self-distillation, FarSLIP employs patch-to-patch distillation to align local and global visual cues, which improves feature discriminability while preserving semantic coherence. Additionally, to effectively utilize region-text supervision, it employs simple CLS token-based region-category alignment rather than explicit patch-level alignment, further enhancing spatial awareness. FarSLIP features improved fine-grained vision-language alignment in RS domain and sets a new state of the art not only on RS open-vocabulary semantic segmentation, but also on image-level tasks such as zero-shot classification and image-text retrieval. Our dataset, code, and models are available at https://github.com/NJU-LHRS/FarSLIP.		Code Link
EBind: a practical approach to space binding	2025-11-18	Show We simplify space binding by focusing on two core components, a single encoder per modality and high-quality data; enabling training state-of-the-art models on a single GPU in a few hours as opposed to multiple days. We present EBind, an Easy, data-centric, and parameter-efficient method to Bind the embedding spaces of multiple contrastive models. We demonstrate that a simple 1.8B-parameter image-text-video-audio-3D model can outperform models 4 to 17x the size. The key to achieving this is a carefully curated dataset of three complementary data sources: i) 6.7M fully-automated multimodal quintuples sourced via SOTA retrieval models, ii) 1M diverse, semi-automated triples annotated by humans as negative, partial, or positive matches, and iii) 3.4M pre-existing captioned data items. We use 13 different evaluations to demonstrate the value of each data source. Due to limitations with existing benchmarks, we further introduce the first high-quality, consensus-annotated zero-shot classification benchmark between audio and PCs. In contrast to related work, we will open-source our code, model weights, and datasets.		None
Insight-A: Attribution-aware for Multimodal Misinformation Detection	2025-11-17	Show AI-generated content (AIGC) technology has emerged as a prevalent alternative to create multimodal misinformation on social media platforms, posing unprecedented threats to societal safety. However, standard prompting leverages multimodal large language models (MLLMs) to identify the emerging misinformation, which ignores the misinformation attribution. To this end, we present Insight-A, exploring attribution with MLLM insights for detecting multimodal misinformation. Insight-A makes two efforts: I) attribute misinformation to forgery sources, and II) an effective pipeline with hierarchical reasoning that detects distortions across modalities. Specifically, to attribute misinformation to forgery traces based on generation patterns, we devise cross-attribution prompting (CAP) to model the sophisticated correlations between perception and reasoning. Meanwhile, to reduce the subjectivity of human-annotated prompts, automatic attribution-debiased prompting (ADP) is used for task adaptation on MLLMs. Additionally, we design image captioning (IC) to achieve visual details for enhancing cross-modal consistency checking. Extensive experiments demonstrate the superiority of our proposal and provide a new paradigm for multimodal misinformation detection in the era of AIGC.		None
Language-Guided Invariance Probing of Vision-Language Models	2025-11-17	Show Recent vision-language models (VLMs) such as CLIP, OpenCLIP, EVA02-CLIP and SigLIP achieve strong zero-shot performance, but it is unclear how reliably they respond to controlled linguistic perturbations. We introduce Language-Guided Invariance Probing (LGIP), a benchmark that measures (i) invariance to meaning-preserving paraphrases and (ii) sensitivity to meaning-changing semantic flips in image-text matching. Using 40k MS COCO images with five human captions each, we automatically generate paraphrases and rule-based flips that alter object category, color or count, and summarize model behavior with an invariance error, a semantic sensitivity gap and a positive-rate statistic. Across nine VLMs, EVA02-CLIP and large OpenCLIP variants lie on a favorable invariance-sensitivity frontier, combining low paraphrase-induced variance with consistently higher scores for original captions than for their flipped counterparts. In contrast, SigLIP and SigLIP2 show much larger invariance error and often prefer flipped captions to the human descriptions, especially for object and color edits. These failures are largely invisible to standard retrieval metrics, indicating that LGIP provides a model-agnostic diagnostic for the linguistic robustness of VLMs beyond conventional accuracy scores.		None
Tracing and Mitigating Hallucinations in Multimodal LLMs via Dynamic Attention Localization	2025-11-17	Show Multimodal Large Language Models (MLLMs) achieve strong performance on tasks like image captioning and visual question answering, but remain prone to hallucinations, where generated text conflicts with the visual input. Prior work links this partly to insufficient visual attention, but existing attention-based detectors and mitigation typically apply uniform adjustments across layers and heads, obscuring where errors originate. In this paper, we first show these methods fail to accurately localize problematic layers. Then, we introduce two diagnostics: Layer Image Attention Entropy (LIAE) which flags anomalous layers, and Image Attention Focus (IAF) which scores attention heads within those layers. Analysis shows that LIAE pinpoints faulty layers and IAF reliably ranks heads that warrant correction. Guided by these signals, we propose Dynamic Layer-wise Entropy and Attention Fusion (D-LEAF), a task-agnostic, attention-guided method that dynamically localizes and corrects errors during inference with negligible overhead. Furthermore, by establishing a connection between D-LEAF and DPO, we provide theoretical justification for the effectiveness of D-LEAF. Results show our D-LEAF delivers a 53% relative improvement on standard captioning benchmarks, and on VQA both accuracy and F1-score improve by approximately 4%, substantially suppressing hallucinations while preserving efficiency.		None
SRD: Reinforcement-Learned Semantic Perturbation for Backdoor Defense in VLMs	2025-11-17	Show Visual language models (VLMs) have made significant progress in image captioning tasks, yet recent studies have found they are vulnerable to backdoor attacks. Attackers can inject undetectable perturbations into the data during inference, triggering abnormal behavior and generating malicious captions. These attacks are particularly challenging to detect and defend against due to the stealthiness and cross-modal propagation of the trigger signals. In this paper, we identify two key vulnerabilities by analyzing existing attack patterns: (1) the model exhibits abnormal attention concentration on certain regions of the input image, and (2) backdoor attacks often induce semantic drift and sentence incoherence. Based on these insights, we propose Semantic Reward Defense (SRD), a reinforcement learning framework that mitigates backdoor behavior without requiring any prior knowledge of trigger patterns. SRD learns to apply discrete perturbations to sensitive contextual regions of image inputs via a deep Q-network policy, aiming to confuse attention and disrupt the activation of malicious paths. To guide policy optimization, we design a reward signal named semantic fidelity score, which jointly assesses the semantic consistency and linguistic fluency of the generated captions, encouraging the agent to achieve a robust yet faithful output. SRD offers a trigger-agnostic, policy-interpretable defense paradigm that effectively mitigates local (TrojVLM) and global (Shadowcast) backdoor attacks, reducing ASR to 3.6% and 5.6% respectively, with less than 15% average CIDEr drop on the clean inputs. Our codes can be found at https://github.com/Ciconey/SRD.git.	AAAI2026	Code Link
Scaling Laws for Conditional Emergence of Multilingual Image Captioning via Generalization from Translation	2025-11-16	Show Cross-lingual, cross-task transfer is challenged by task-specific data scarcity, which becomes more severe as language support grows and is further amplified in vision-language models (VLMs). We investigate multilingual generalization in encoder-decoder transformer VLMs to enable zero-shot image captioning in languages encountered only in the translation task. In this setting, the encoder must learn to generate generalizable, task-aware latent vision representations to instruct the decoder via inserted cross-attention layers. To analyze scaling behavior, we train Florence-2 based and Gemma-2 based models (0.4B to 11.2B parameters) on a synthetic dataset using varying compute budgets. While all languages in the dataset have image-aligned translations, only a subset of them include image captions. Notably, we show that captioning can emerge using a language prefix, even when this language only appears in the translation task. We find that indirect learning of unseen task-language pairs adheres to scaling laws that are governed by the multilinguality of the model, model size, and seen training samples. Finally, we demonstrate that the scaling laws extend to downstream tasks, achieving competitive performance through fine-tuning in multimodal machine translation (Multi30K, CoMMuTE), lexical disambiguation (CoMMuTE), and image captioning (Multi30K, XM3600, COCO Karpathy).		None
DenseAnnotate: Enabling Scalable Dense Caption Collection for Images and 3D Scenes via Spoken Descriptions	2025-11-16	Show With the rapid adoption of multimodal large language models (MLLMs) across diverse applications, there is a pressing need for task-centered, high-quality training data. A key limitation of current training datasets is their reliance on sparse annotations mined from the Internet or entered via manual typing that capture only a fraction of an image's visual content. Dense annotations are more valuable but remain scarce. Traditional text-based annotation pipelines are poorly suited for creating dense annotations: typing limits expressiveness, slows annotation speed, and underrepresents nuanced visual features, especially in specialized areas such as multicultural imagery and 3D asset annotation. In this paper, we present DenseAnnotate, an audio-driven online annotation platform that enables efficient creation of dense, fine-grained annotations for images and 3D assets. Annotators narrate observations aloud while synchronously linking spoken phrases to image regions or 3D scene parts. Our platform incorporates speech-to-text transcription and region-of-attention marking. To demonstrate the effectiveness of DenseAnnotate, we conducted case studies involving over 1,000 annotators across two domains: culturally diverse images and 3D scenes. We curate a human-annotated multi-modal dataset of 3,531 images, 898 3D scenes, and 7,460 3D objects, with audio-aligned dense annotations in 20 languages, including 8,746 image captions, 2,000 scene captions, and 19,000 object captions. Models trained on this dataset exhibit improvements of 5% in multilingual, 47% in cultural alignment, and 54% in 3D spatial capabilities. Our results show that our platform offers a feasible approach for future vision-language research and can be applied to various tasks and diverse types of data.		None
Studying Illustrations in Manuscripts: An Efficient Deep-Learning Approach	2025-11-15	Show The recent Artificial Intelligence (AI) revolution has opened transformative possibilities for the humanities, particularly in unlocking the visual content embedded in historical manuscripts. While digital archives now offer unprecedented access to these materials, the ability to systematically study illustrations at a large scale remains challenging. Our study presents a fast and scalable AI approach for detecting, extracting, and describing illustrations in digitized manuscripts. Focusing on collections like the Vatican Library, our system enables efficient visual analysis across millions of pages. Our pipeline consists of three stages: (1) a fine-tuned image classification model filters out text-only pages; (2) an efficient object detection model identifies and crops illustrations; and (3) a multimodal image captioning model generates concise, human-readable descriptions. These are stored in a searchable database, allowing scholars to retrieve relevant visual materials through keyword queries. By harnessing the power of recent AI advancements, we enable large-scale visual research that was previously impractical, empowering scholars in historical studies, art history, and cultural heritage to explore visual motifs, artistic styles, and cross-cultural influences with new precision and speed. Applying our pipeline to over three million digitized manuscript pages, we automatically identified and extracted more than 200,000 unique illustrations. This scale of processing in under 0.06 seconds per page, dramatically outperforms traditional segmentation techniques in both efficiency and accessibility for visual scholarship. Our work demonstrates how cutting-edge AI tools can profoundly reshape scholarly workflows and open new avenues for multidisciplinary research in the age of digital manuscripts.	14 pages, 5 figures	None
Landsat30-AU: A Vision-Language Dataset for Australian Landsat Imagery	2025-11-15	Show Vision language models (VLMs) that enable natural language interaction with satellite imagery can democratize Earth observation by accelerating expert workflows, making data accessible to non-specialists, and enabling planet-scale automation. However, existing datasets focus mainly on short-term, high-resolution imagery from a limited number of satellites, overlooking low-resolution, multi-satellite, long-term archives, such as Landsat, that are essential for affordable and bias-robust global monitoring. We address this gap with Landsat30-AU, a large-scale vision-language dataset built from 30-meter resolution imagery collected by four Landsat satellites (5, 7, 8, and 9) over Australia, spanning more than 36 years. The dataset includes two components: Landsat30-AU-Cap, containing $196,262$ image-caption pairs, and Landsat30-AU-VQA, comprising 17,725 human-verified visual question answering (VQA) samples across eight remote sensing domains. Both datasets are curated through a bootstrapped pipeline that leverages generic VLMs with iterative refinement and human verification to ensure quality. Our evaluation of eight VLMs on our benchmark reveals that off-the-shelf models struggle to understand satellite imagery. The open-source remote-sensing VLM EarthDial achieves only 0.07 SPIDEr in captioning and a VQA accuracy of 0.48, highlighting the limitations of current approaches. Encouragingly, lightweight fine-tuning of Qwen2.5-VL-7B on Landsat30-AU improves captioning performance from 0.11 to 0.31 SPIDEr and boosts VQA accuracy from 0.74 to 0.87. Code and data are available at https://github.com/papersubmit1/landsat30-au.		Code Link
SpaceVLM: Sub-Space Modeling of Negation in Vision-Language Models	2025-11-15	Show Vision-Language Models (VLMs) struggle with negation. Given a prompt like "retrieve (or generate) a street scene without pedestrians," they often fail to respect the "not." Existing methods address this limitation by fine-tuning on large negation datasets, but such retraining often compromises the model's zero-shot performance on affirmative prompts. We show that the embedding space of VLMs, such as CLIP, can be divided into semantically consistent subspaces. Based on this property, we propose a training-free framework that models negation as a subspace in the joint embedding space rather than a single point (Figure 1). To find the matching image for a caption such as "A but not N," we construct two spherical caps around the embeddings of A and N, and we score images by the central direction of the region that is close to A and far from N. Across retrieval, MCQ, and text-to-image tasks, our method improves negation understanding by about 30% on average over prior methods. It closes the gap between affirmative and negated prompts while preserving the zero-shot performance that fine-tuned models fail to maintain. Code will be released upon publication.		None
Suppressing VLM Hallucinations with Spectral Representation Filtering	2025-11-15	Show Vision-language models (VLMs) frequently produce hallucinations in the form of descriptions of objects, attributes, or relations that do not exist in the image due to over-reliance on language priors and imprecise cross-modal grounding. We introduce Spectral Representation Filtering (SRF), a lightweight, training-free method to suppress such hallucinations by analyzing and correcting the covariance structure of the model's representations. SRF identifies low-rank hallucination modes through eigendecomposition of the covariance of the differences between features collected for truthful and hallucinatory captions, revealing structured biases in the feature space. A soft spectral filter then attenuates these modes in the feed-forward projection weights of deeper vLLM layers, equalizing feature variance while preserving semantic fidelity. Unlike decoding or retraining-based approaches, SRF operates entirely post-hoc, incurs zero inference overhead, and requires no architectural modifications. Across three families of VLMs (LLaVA-1.5, MiniGPT-4, and mPLUG-Owl2), SRF consistently reduces hallucination rates on MSCOCO, POPE-VQA, and other visual tasks benchmarks, achieving state-of-the-art faithfulness without degrading caption quality.		None
MicroVQA++: High-Quality Microscopy Reasoning Dataset with Weakly Supervised Graphs for Multimodal Large Language Model	2025-11-14	Show Multimodal Large Language Models are increasingly applied to biomedical imaging, yet scientific reasoning for microscopy remains limited by the scarcity of large-scale, high-quality training data. We introduce MicroVQA++, a three-stage, large-scale and high-quality microscopy VQA corpus derived from the BIOMEDICA archive. Stage one bootstraps supervision from expert-validated figure-caption pairs sourced from peer-reviewed articles. Stage two applies HiCQA-Graph, a novel heterogeneous graph over images, captions, and QAs that fuses NLI-based textual entailment, CLIP-based vision-language alignment, and agent signals to identify and filter inconsistent samples. Stage three uses a MultiModal Large Language Model (MLLM) agent to generate multiple-choice questions (MCQ) followed by human screening. The resulting release comprises a large training split and a human-checked test split whose Bloom's level hard-sample distribution exceeds the MicroVQA benchmark. Our work delivers (i) a quality-controlled dataset that couples expert literature with graph-based filtering and human refinement; (ii) HiCQA-Graph, the first graph that jointly models (image, caption, QA) for cross-modal consistency filtering; (iii) evidence that careful data construction enables 4B-scale MLLMs to reach competitive microscopy reasoning performance (e.g., GPT-5) and achieve state-of-the-art performance among open-source MLLMs. Code and dataset will be released after the review process concludes.	11 pages, 4 figures	None
Discovering Meaningful Units with Visually Grounded Semantics from Image Captions	2025-11-14	Show Fine-grained knowledge is crucial for vision-language models to obtain a better understanding of the real world. While there has been work trying to acquire this kind of knowledge in the space of vision and language, it has mostly focused on aligning the image patches with the tokens on the language side. However, image patches do not have any meaning to the human eye, and individual tokens do not necessarily carry groundable information in the image. It is groups of tokens which describe different aspects of the scene. In this work, we propose a model which groups the caption tokens as part of its architecture in order to capture a fine-grained representation of the language. We expect our representations to be at the level of objects present in the image, and therefore align our representations with the output of an image encoder trained to discover objects. We show that by learning to group the tokens, the vision-language model has a better fine-grained understanding of vision and language. In addition, the token groups that our model discovers are highly similar to groundable phrases in text, both qualitatively and quantitatively.		None
ORIC: Benchmarking Object Recognition under Contextual Incongruity in Large Vision-Language Models	2025-11-14	Show Large Vision-Language Models (LVLMs) excel at captioning, visual question answering, and robotics by combining vision and language, yet they often miss obvious objects or hallucinate nonexistent ones in atypical scenes. We examine these failures through the lens of uncertainty, focusing on contextual incongruity, where objects appear unexpectedly or fail to appear in expected contexts, and show that such cases increase recognition difficulty for state-of-the-art LVLMs. To study this regime, we introduce the Object Recognition in Incongruous Context (ORIC) framework, which constructs incongruous object-context pairs through two complementary strategies: (1) LLM-guided sampling to identify hard-to-recognize objects present in the image and (2) CLIP-guided sampling to mine plausible but absent ones. Applied to MSCOCO, ORIC produces ORIC-Bench and ORIC-style training data. Evaluating 18 LVLMs and 2 open-vocabulary detectors reveals substantial performance drops and bias patterns under incongruous contexts. Fine-tuning Qwen3-VL-8B-Instruct with Visual Reinforcement Fine-Tuning on 600 ORIC-style samples improves results on ORIC-Bench, AMBER, and HallusionBench. Overall, we show that contextual incongruity is a key source of uncertainty and provide tools for more reliable LVLMs. The code is available at https://github.com/ZhaoyangLi-1/ORIC.		Code Link
Draft and Refine with Visual Experts	2025-11-14	Show While recent Large Vision-Language Models (LVLMs) exhibit strong multimodal reasoning abilities, they often produce ungrounded or hallucinated responses because they rely too heavily on linguistic priors instead of visual evidence. This limitation highlights the absence of a quantitative measure of how much these models actually use visual information during reasoning. We propose Draft and Refine (DnR), an agent framework driven by a question-conditioned utilization metric. The metric quantifies the model's reliance on visual evidence by first constructing a query-conditioned relevance map to localize question-specific cues and then measuring dependence through relevance-guided probabilistic masking. Guided by this metric, the DnR agent refines its initial draft using targeted feedback from external visual experts. Each expert's output (such as boxes or masks) is rendered as visual cues on the image, and the model is re-queried to select the response that yields the largest improvement in utilization. This process strengthens visual grounding without retraining or architectural changes. Experiments across VQA and captioning benchmarks show consistent accuracy gains and reduced hallucination, demonstrating that measuring visual utilization provides a principled path toward more interpretable and evidence-driven multimodal agent systems.		None
EmoVid: A Multimodal Emotion Video Dataset for Emotion-Centric Video Understanding and Generation	2025-11-14	Show Emotion plays a pivotal role in video-based expression, but existing video generation systems predominantly focus on low-level visual metrics while neglecting affective dimensions. Although emotion analysis has made progress in the visual domain, the video community lacks dedicated resources to bridge emotion understanding with generative tasks, particularly for stylized and non-realistic contexts. To address this gap, we introduce EmoVid, the first multimodal, emotion-annotated video dataset specifically designed for creative media, which includes cartoon animations, movie clips, and animated stickers. Each video is annotated with emotion labels, visual attributes (brightness, colorfulness, hue), and text captions. Through systematic analysis, we uncover spatial and temporal patterns linking visual features to emotional perceptions across diverse video forms. Building on these insights, we develop an emotion-conditioned video generation technique by fine-tuning the Wan2.1 model. The results show a significant improvement in both quantitative metrics and the visual quality of generated videos for text-to-video and image-to-video tasks. EmoVid establishes a new benchmark for affective video computing. Our work not only offers valuable insights into visual emotion analysis in artistically styled videos, but also provides practical methods for enhancing emotional expression in video generation.	15 pa... 15 pages, 12 figures. Accepted as an Oral presentation at AAAI 2026. For code and dataset, see https://zane-zyqiu.github.io/EmoVid	Code Link
MOSABench: Multi-Object Sentiment Analysis Benchmark for Evaluating Multimodal Large Language Models Understanding of Complex Image	2025-11-14	Show Multimodal large language models (MLLMs) have shown remarkable progress in high-level semantic tasks such as visual question answering, image captioning, and emotion recognition. However, despite advancements, there remains a lack of standardized benchmarks for evaluating MLLMs performance in multi-object sentiment analysis, a key task in semantic understanding. To address this gap, we introduce MOSABench, a novel evaluation dataset designed specifically for multi-object sentiment analysis. MOSABench includes approximately 1,000 images with multiple objects, requiring MLLMs to independently assess the sentiment of each object, thereby reflecting real-world complexities. Key innovations in MOSABench include distance-based target annotation, post-processing for evaluation to standardize outputs, and an improved scoring mechanism. Our experiments reveal notable limitations in current MLLMs: while some models, like mPLUG-owl and Qwen-VL2, demonstrate effective attention to sentiment-relevant features, others exhibit scattered focus and performance declines, especially as the spatial distance between objects increases. This research underscores the need for MLLMs to enhance accuracy in complex, multi-object sentiment analysis tasks and establishes MOSABench as a foundational tool for advancing sentiment analysis capabilities in MLLMs.		None
LISA: A Layer-wise Integration and Suppression Approach for Hallucination Mitigation in Multimodal Large Language Models	2025-11-13	Show Multimodal Large Language Models (MLLMs) excel in vision-language tasks such as image captioning but remain prone to object hallucinations, where they describe objects that do not appear in the image. To mitigate this, we propose LISA, a Layer-wise Integration and Suppression Approach. LISA leverages the layer-wise functional roles in MLLMs: shallow layers provide visual grounding, middle layers encode semantics, and deep layers tend to amplify spurious signals. First, layer-wise spectral modulation stabilizes attention by suppressing over-amplified activations in deeper layers while preserving alignment cues in earlier layers. Second, token-level logits from selected layers are fused via anchor-based routing, with token-wise anchor selection and soft logit fusion enabling adaptive integration during decoding. LISA is fully plug-and-play and can be seamlessly integrated into existing MLLMs, including Qwen2.5-VL. Experiments on multiple benchmarks show that LISA reduces hallucinations by up to 53.6% in $\text{CHAIR}_\text{I}$ and improves POPE F1 by up to 5.1%, demonstrating strong generalization across models and tasks. Our code is available at https://github.com/zhlisa1010-eng/LISA.		Code Link
Remodeling Semantic Relationships in Vision-Language Fine-Tuning	2025-11-13	Show Vision-language fine-tuning has emerged as an efficient paradigm for constructing multimodal foundation models. While textual context often highlights semantic relationships within an image, existing fine-tuning methods typically overlook this information when aligning vision and language, thus leading to suboptimal performance. Toward solving this problem, we propose a method that can improve multimodal alignment and fusion based on both semantics and relationships.Specifically, we first extract multilevel semantic features from different vision encoder to capture more visual cues of the relationships. Then, we learn to project the vision features to group related semantics, among which are more likely to have relationships. Finally, we fuse the visual features with the textual by using inheritable cross-attention, where we globally remove the redundant visual relationships by discarding visual-language feature pairs with low correlation. We evaluate our proposed method on eight foundation models and two downstream tasks, visual question answering and image captioning, and show that it outperforms all existing methods.		None
LayerPeeler: Autoregressive Peeling for Layer-wise Image Vectorization	2025-11-13	Show Image vectorization is a powerful technique that converts raster images into vector graphics, enabling enhanced flexibility and interactivity. However, popular image vectorization tools struggle with occluded regions, producing incomplete or fragmented shapes that hinder editability. While recent advancements have explored optimization-based and learning-based layer-wise image vectorization, these methods face limitations in vectorization quality and flexibility. In this paper, we introduce LayerPeeler, a novel layer-wise image vectorization approach that addresses these challenges through a progressive simplification paradigm. The key to LayerPeeler's success lies in its autoregressive peeling strategy: by identifying and removing the topmost non-occluded layers while recovering underlying content, we generate vector graphics with complete paths and coherent layer structures. Our method leverages vision-language models to construct a layer graph that captures occlusion relationships among elements, enabling precise detection and description for non-occluded layers. These descriptive captions are used as editing instructions for a finetuned image diffusion model to remove the identified layers. To ensure accurate removal, we employ localized attention control that precisely guides the model to target regions while faithfully preserving the surrounding content. To support this, we contribute a large-scale dataset specifically designed for layer peeling tasks. Extensive quantitative and qualitative experiments demonstrate that LayerPeeler significantly outperforms existing techniques, producing vectorization results with superior path semantics, geometric regularity, and visual fidelity.	Proje... Project Page: https://layerpeeler.github.io/	None
RangeSAM: On the Potential of Visual Foundation Models for Range-View represented LiDAR segmentation	2025-11-13	Show Point cloud segmentation is central to autonomous driving and 3D scene understanding. While voxel- and point-based methods dominate recent research due to their compatibility with deep architectures and ability to capture fine-grained geometry, they often incur high computational cost, irregular memory access, and limited real-time efficiency. In contrast, range-view methods, though relatively underexplored - can leverage mature 2D semantic segmentation techniques for fast and accurate predictions. Motivated by the rapid progress in Visual Foundation Models (VFMs) for captioning, zero-shot recognition, and multimodal tasks, we investigate whether SAM2, the current state-of-the-art VFM for segmentation tasks, can serve as a strong backbone for LiDAR point cloud segmentation in the range view. We present , to our knowledge, the first range-view framework that adapts SAM2 to 3D segmentation, coupling efficient 2D feature extraction with standard projection/back-projection to operate on point clouds. To optimize SAM2 for range-view representations, we implement several architectural modifications to the encoder: (1) a novel module that emphasizes horizontal spatial dependencies inherent in LiDAR range images, (2) a customized configuration of tailored to the geometric properties of spherical projections, and (3) an adapted mechanism in the encoder backbone specifically designed to capture the unique spatial patterns and discontinuities present in range-view pseudo-images. Our approach achieves competitive performance on SemanticKITTI while benefiting from the speed, scalability, and deployment simplicity of 2D-centric pipelines. This work highlights the viability of VFMs as general-purpose backbones for 3D perception and opens a path toward unified, foundation-model-driven LiDAR segmentation. Results lets us conclude that range-view segmentation methods using VFMs leads to promising results.		None
Caption, Create, Continue: Continual Learning with Pre-trained Generative Vision-Language Models	2025-11-13	Show Continual learning (CL) enables models to adapt to evolving data streams without catastrophic forgetting, a fundamental requirement for real-world AI systems. However, the current methods often depend on large replay buffers or heavily annotated datasets which are impractical due to storage, privacy, and cost constraints. We propose CLTS (Continual Learning via Text-Image Synergy), a novel class-incremental framework that mitigates forgetting without storing real task data. CLTS leverages pre-trained vision-language models, BLIP (Bootstrapping Language-Image Pre-training) for caption generation and stable diffusion for sample generation. Each task is handled by a dedicated Task Head, while a Task Router learns to assign inputs to the correct Task Head using the generated data. On three benchmark datasets, CLTS improves average task accuracy by up to 54% and achieves 63 times better memory efficiency compared to four recent continual learning baselines, demonstrating improved retention and adaptability. CLTS introduces a novel perspective by integrating generative text-image augmentation for scalable continual learning.	This ... This is the revised and peer-reviewed version of our paper, accepted and published in the Proceedings of the 34th ACM International Conference on Information and Knowledge Management (CIKM 2025)	None
Anomagic: Crossmodal Prompt-driven Zero-shot Anomaly Generation	2025-11-13	Show We propose Anomagic, a zero-shot anomaly generation method that produces semantically coherent anomalies without requiring any exemplar anomalies. By unifying both visual and textual cues through a crossmodal prompt encoding scheme, Anomagic leverages rich contextual information to steer an inpainting-based generation pipeline. A subsequent contrastive refinement strategy enforces precise alignment between synthesized anomalies and their masks, thereby bolstering downstream anomaly detection accuracy. To facilitate training, we introduce AnomVerse, a collection of 12,987 anomaly-mask-caption triplets assembled from 13 publicly available datasets, where captions are automatically generated by multimodal large language models using structured visual prompts and template-based textual hints. Extensive experiments demonstrate that Anomagic trained on AnomVerse can synthesize more realistic and varied anomalies than prior methods, yielding superior improvements in downstream anomaly detection. Furthermore, Anomagic can generate anomalies for any normal-category image using user-defined prompts, establishing a versatile foundation model for anomaly generation.		None
Mitigating Perception Bias: A Training-Free Approach to Enhance LMM for Image Quality Assessment	2025-11-13	Show Despite the impressive performance of large multimodal models (LMMs) in high-level visual tasks, their capacity for image quality assessment (IQA) remains limited. One main reason is that LMMs are primarily trained for high-level tasks (e.g., image captioning), emphasizing unified image semantics extraction under varied quality. Such semantic-aware yet quality-insensitive perception bias inevitably leads to a heavy reliance on image semantics when those LMMs are forced for quality rating. In this paper, instead of retraining or tuning an LMM costly, we propose a training-free debiasing framework, in which the image quality prediction is rectified by mitigating the bias caused by image semantics. Specifically, we first explore several semantic-preserving distortions that can significantly degrade image quality while maintaining identifiable semantics. By applying these specific distortions to the query or test images, we ensure that the degraded images are recognized as poor quality while their semantics mainly remain. During quality inference, both a query image and its corresponding degraded version are fed to the LMM along with a prompt indicating that the query image quality should be inferred under the condition that the degraded one is deemed poor quality. This prior condition effectively aligns the LMM's quality perception, as all degraded images are consistently rated as poor quality, regardless of their semantic variance. Finally, the quality scores of the query image inferred under different prior conditions (degraded versions) are aggregated using a conditional probability model. Extensive experiments on various IQA datasets show that our debiasing framework could consistently enhance the LMM performance.		None
Regional Attention-Enhanced Swin Transformer for Clinically Relevant Medical Image Captioning	2025-11-13	Show Automated medical image captioning translates complex radiological images into diagnostic narratives that can support reporting workflows. We present a Swin-BART encoder-decoder system with a lightweight regional attention module that amplifies diagnostically salient regions before cross-attention. Trained and evaluated on ROCO, our model achieves state-of-the-art semantic fidelity while remaining compact and interpretable. We report results as mean$\pm$std over three seeds and include $95%$ confidence intervals. Compared with baselines, our approach improves ROUGE (proposed 0.603, ResNet-CNN 0.356, BLIP2-OPT 0.255) and BERTScore (proposed 0.807, BLIP2-OPT 0.645, ResNet-CNN 0.623), with competitive BLEU, CIDEr, and METEOR. We further provide ablations (regional attention on/off and token-count sweep), per-modality analysis (CT/MRI/X-ray), paired significance tests, and qualitative heatmaps that visualize the regions driving each description. Decoding uses beam search (beam size $=4$), length penalty $=1.1$, $no_repeat_ngram_size$ $=3$, and max length $=128$. The proposed design yields accurate, clinically phrased captions and transparent regional attributions, supporting safe research use with a human in the loop.		None
Edit Flows: Flow Matching with Edit Operations	2025-11-12	Show Autoregressive generative models naturally generate variable-length sequences, while non-autoregressive models struggle, often imposing rigid, token-wise structures. We propose Edit Flows, a non-autoregressive model that overcomes these limitations by defining a discrete flow over sequences through edit operations$\unicode{x2013}$insertions, deletions, and substitutions. By modeling these operations within a Continuous-time Markov Chain over the sequence space, Edit Flows enable flexible, position-relative generation that aligns more closely with the structure of sequence data. Our training method leverages an expanded state space with auxiliary variables, making the learning process efficient and tractable. Empirical results show that Edit Flows outperforms both autoregressive and mask models on image captioning and significantly outperforms the mask construction in text and code generation.		None