🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

We will present this work at ICML in person next Thursday. Feel free to drop by if you have any questions or just want to say hello! #ICML2024

I had an amazing internship experience working at the Discovery team! Don't hesitate to reach out/apply if you are passionate about RL, Planning and LLMs!

Can robots self-improve by collecting data autonomously🤖? Introducing SOAR: a system for large-scale autonomous data collection 🚀 and autonomous improvement📈of a multi-task language-conditioned policy in diverse scenes without human interventions . auto-improvement.github.io

I'm recruiting PhD students this year with interest in machine learning, embodied AI, or AI for science! If you are interested in constructing fundamental tools to improve Generative AI and exploring how these tools can be used for intelligent embodied agents and science,

Thrilled to share that I’ll be joining Google DeepMind as a Research Scientist with the Discovery Team! It’s a dream come true—8 years ago, I watched AlphaGo live in high school, and now I am lucky enough to be part of this incredible journey. Can’t wait to discover what’s next!

We will present EPO: Hierarchical LLM Agents with Environment Preference Optimization at #EMNLP2024 next week, which achieves SOTA performance on ALFRED by finetuning a hierarchical Llama2-based agent and modeling the environment's preference signals. Check out the thread below!

Natural Language Reinforcement Learning (NLRL) redefines Reinforcement Learning (RL). The main idea: In NLRL, the core parts of RL like goals, strategies, and evaluation methods are reimagined using natural language instead of rigid math. What are the benefits? - NLRL uses not

Frozen diffusion models pretrained on web images and videos can teach embodied agents to accomplish novel behaviors and goals in diverse physical environments! We present TADPoLe🐸, a way to “distill” large-scale pretraining into text-conditioned policy learning! #NeurIPS2024

Want to know how web-scale pretrained diffusion models can provide text-conditioned reward signals for learning novel behaviors across different robotic configurations and environments? Check our poster at #NeurIPS2024 on Thursday, December 12th at 4:30 pm (Poster #6706)!

NLRL is now open-sourced at github.com/waterhorse1/Na…! Check it and welcome to try NLRL in other LLM agents, reasoning, and coding tasks!

NLRL, or Natural Language Reinforcement Learning, is about adapting RL methods to work in the natural language field. Traditional RL aims to learn a policy (strategy) guiding the agent to the best action in each state. Instead of this, NLRL integrates a Chain-of-Thought

🚀 DeepSeek-R1 is here! ⚡ Performance on par with OpenAI-o1 📖 Fully open-source model & technical report 🏆 MIT licensed: Distill & commercialize freely! 🌐 Website & API are live now! Try DeepThink at chat.deepseek.com today! 🐋 1/n

In this demo, Ultra Mobile Vehicle (UMV) drives, turns, jumps, tricks, and comes to a sudden stop called a track-stand. All of the driving, landings, balance, and track-stands are done using reinforcement learning.

Can we make robot policy evaluation easier and less time consuming? Introducing AutoEval, a system that *autonomously* evaluates generalist policies 24/7 and closely matches human results. We make 4 tasks 💫publicly available💫 Submit your policy at auto-eval.github.io! 🧵👇

Internet-scale videos capture rich knowledge about motions and behaviors, but how can we effectively adapt them for domain-specific robotic tasks and further unlock behavior generalization? Come to our poster (#418) at #ICLR2025 and chat with Calvin Luo on April 25th at 3:30pm!

Sigh, it's a bit of a mess. Let me just give you guys the full nuance in one stream of consciousness since I think we'll continue to get partial interpretations that confuse everyone. All the little things I post need to always be put together in one place. First, I have long