🎉 Announcing the first Open Science for Foundation Models (SCI-FM) Workshop at #ICLR2025! Join us in advancing transparency and reproducibility in AI through open foundation models. 🤝 Looking to contribute? Join our Program Committee: bit.ly/4acBBjF 🔍 Learn more at:

IMO this is another good work showing the paradigm of “learning = inference + long-term memory” (Previous one was HippoRAG led by Bernal Jiménez on semantic knowledge.) Here long-term memory is procedural knowledge organized as Python APIs. Such APIs allow reliable

WTF just happened 🤣🤣

Will be at ICLR from April 24-28. Can't wait to see my old/new friends! Also, please reach out if you wanna discuss anything about research in agents! #ICLR2025

question to RL people: why the reward in RL has to be numerical? is it a design by nature? or is it mainly an expedient design to simplify the model? eager to learn about your opinions

Project DeepWiki Up-to-date documentation you can talk to, for every repo in the world. Think Deep Research for GitHub – powered by Devin. It’s free for open-source, no sign-up! Visit deepwiki com or just swap github → deepwiki on any repo URL:

Almost all my knowledge about robotics comes from Luke!

Good scientists have a deep psychological need for crisp definitions and self-consistent models of the world. Most people are comfortable holding worldviews that are fragmented, inconsistent, and fluid, where one's degree of belief in various things fluctuates based on context

Tired of editing methods that require training, handcrafted subjects, or external memory? 🚀 #UltraEdit — Training-, subject-, and memory-free, for Lifelong Model Editing Compare to the prior best ✅New SOTA on 4 datasets and 6 models 🏎️7× faster – 20K samples within 5 mins on a

🚀 Thrilled to unveil the most exciting project of my PhD: Explorer — Scaling Exploration-driven Web Trajectory Synthesis for Multimodal Web Agents TL;DR: A scalable multi-agent pipeline that leverages exploration for diverse web agent trajectory synthesis. 📄 Paper:

⁉️Can you really trust Computer-Use Agents (CUAs) to control your computer⁉️ Not yet, Anthropic Opus 4 shows an alarming 48% Attack Success Rate against realistic internet injection❗️ Introducing RedTeamCUA: realistic, interactive, and controlled sandbox environments for

Zihao has been aware of this for three months. There's a more generalized claim underlying recent assertions about RL (in the context of using Qwen for math) like you can do RL with one example/internal rewards/spurious rewards: for Qwen on math, you just don't need RL at all!