personal website

Open sourced predictions, execution logs, trajectories, and results from model inference + evaluation runs on the SWE-bench task.

Post-training with Tinker

SkyRL: A Modular Full-stack RL Library for LLMs adapted to allow for detection multi-turn multi-agent RL

A scalable asynchronous reinforcement learning implementation with in-flight weight updates.

Democratizing Reinforcement Learning for LLMs

SWE-bench [Multimodal]: Can Language Models Resolve Real-world Github Issues?

This is the official repository for M2UGen

Simulating Large-Scale Multi-Agent Interactions with Limited Multimodal Senses and Physical Needs

A Jekyll version of the "Forty" theme by HTML5 UP.

🐟 A simple theme for Jekyll. Live at https://eliottvincent.github.io/bay/

Dopamine is a research framework for fast prototyping of reinforcement learning algorithms.

A Deep RL Wordle Bot