diff --git a/index.md b/index.md index f0e4d52..af2b241 100644 --- a/index.md +++ b/index.md @@ -11,9 +11,9 @@ order: 0
-

LM-class is an education resource for contemporary language modeling, broadly construed. The materials were developed for Cornell CS 5740 Natural Language Processing, and were first used in Spring 2024. This class constituted a significant revision of CS 5740. The target audience of the class is MEng/MS graduate students. The material is likely also appropriate for senior undergraduates. While the technical material is a good fit for PhDs and researchers, this class does not focus on research, and is designed primarily for practitioners.

+

LM-class is an introduction-level education resource for contemporary language modeling, broadly construed. It relies on prior understanding of machine learning and neural networks at the introduction level, as well as undergraduate-level programming, probability theory, and linear algebra. The target audience of the class is MEng/MS graduate students. The material is likely also appropriate for senior undergraduates. While the technical material is a good fit for PhDs and researchers, this class does not focus on research, and is designed primarily for practitioners.

The materials include lectures and assignments. The class emphasizes technical depth rather than coverage. It does not aim to provide a broad overview of everything that is happening in the field. The objective is to give students a strong base and the tools to expand their knowledge and update it on their own.

-

LM-class was created by Yoav Artzi, Anne Wu, and Omer Gul. Much of the material was adapted or inspired by existing NLP classes. Each lecture and assignment includes a slide at the end with acknowledgements. If I missed any attribution, I am really sorry. Please let me know, so I can correct. Acknowledgements are listed below.

+

The materials were developed for Cornell CS 5740 Natural Language Processing, and were first used in Spring 2024. This class constituted a significant revision of CS 5740. LM-class was created by Yoav Artzi, Anne Wu, and Omer Gul. Much of the material was adapted or inspired by existing NLP classes. Each lecture and assignment includes a slide at the end with acknowledgements. If I missed any attribution, I am really sorry. Please let me know, so I can correct. Acknowledgements are listed below.

The materials are distributed under the CC BY-NC 4.0 license, and we hope they will find broad and diverse use.

@@ -39,6 +39,10 @@ I am also considering adding a PyTorch tutorial, with focus on: 3. Conducting parameter sweeps 4. Basic debugging techniques +## Beyond LM-class + +LM-class is a considerable revamp of how I teach language modeling and NLP. I am still trying to figure out how it integrates into a broader curriculum, beyond the prerequisites listed above. Currently, the most attractive next step in the sequence where LM-class is the first step is [Stanford's CS 336 Language Modeling from Scratch](https://stanford-cs336.github.io/), which was designed by [Tatsunori Hashimoto](https://thashim.github.io/) and [Percy Liang](https://cs.stanford.edu/~pliang/). CS 336 offers an advanced deep technical dive into training a contemporary language model from scratch. + ## Acknowledgements This resource could not have been created if not for the amazing breadth and richness of teaching material that has been made publicly available by members of the community.