🤔 How to extract knowledge from LLMs to train better RL agents?

📚 Our new paper (w. Q. Zheng, @mikaelhenaff.bsky.social, A. Zhang, A. Grover) studies LLM-driven feedback for NetHack!

Paper: arxiv.org/abs/2410.23022
Code: github.com/facebookrese...

Our visiting students usually publish at ICML, ICLR, and NeurIPS.

My website gives an overview of the kind of research we do:

bamos.github.io

For more of a visual overview, my presentations are also available here:

bamos.github.io/presentations/

Here's my take on an LLM intro for our ML course at Cornell Tech (with V. Kuleshov). It starts with the foundations then goes through code to use pretrained models for 1) generation, 2) chat, and 3) code infilling.

pdf: drive.google.com/file/d/1CTGo...
ipynb: github.com/kuleshov/cor...

here's the \mathbbmss{X} version

hi everyone!! let's try this optimal transport again 🙃