PhD student at MIT CSAIL. Formerly ML @reverielabs, @harvardnlp and @google-research.
world-models. Jupyter Notebook
220lilo. LILO: Library Induction with Language Observations
98self-steering. [COLM 2025] Official repo for Self-Steering Language Models
26battleship. Official repo for Shoot First, Ask Questions Later?
24adversarial-vqa. Jupyter Notebook
12ControlNet. Let us control diffusion models!
5VQA-Counterexamples. Visual Question Answering in Pytorch
2Super-Mario-RL. Reinforcement learning for Super Mario Bros. in OpenAI Gym environment.
2stitch_bindings. Bindings for stitch
1Q-learning. Implements a gain-loss Q-learning model.
1stone. A minimalist, set-theoretic programming language
1