Excited to present our 𝗦𝗹𝗼𝘁 𝗦𝘁𝗮𝘁𝗲 𝗦𝗽𝗮𝗰𝗲 𝗠𝗼𝗱𝗲𝗹𝘀 at #NeurIPS2024! Interested in Modular Spatial-Temporal Representation Learning and its potential to enhance Visual Reasoning? Don’t miss our poster on Fri, Dec 13, 4:30-7:30 PM, East Exhibit Hall A-C (#1811).

🚀 Funded PhD Studentship Alert! 🚀 Join Oxford Applied AI Lab in our mission to advance World Models for robotics and beyond. Two opportunities starting Oct 2025. Deadlines in January 2025. Details in 🧵… Oxford Robotics Institute 🤖 #Robotics #ArtificialInteligence #MachineLearning

My PhD thesis entitled "Generative Flow Networks: Theory and Applications to Structure Learning" is now available on Arxiv 🎓 📖 arxiv.org/abs/2501.05498 🔖 Want to learn what GFlowNets are? Check out Chapters 2, 3 & 4!

We are hiring a PhD intern at NVIDIA research. If you are interested, email me your CV. Research topics: SSM/Mamba, other alternative architectures for LLMs/multimodal LLMs, or more interesting topic than these :)

Yoshua Bengio says when OpenAI develop superintelligent AI they won't share it with the world, but instead will use it to dominate and wipe out other companies and the economies of other countries

I am happy to share our new paper, Monte Carlo Tree Diffusion (MCTD) for System 2 Planning. Unsupervised RL with generative model is one of the remedies to achieve the general agent, then what if applying test-time computing to there? We studied this on Diffusion Planner.

Cooking a new ML model while gazing at East Sea of S. Korea

I’ll be in the NYC area the week of April 14—feel free to reach out if you’d like to catch up! 🇰🇷✈️🇺🇸

Bringing hierarchical structure for search-based inference in LLM. Sharing our recent work "Search-Based Correction of Reasoning Chains for LMs". Work done with Jean-Pierre Falet, Oliver Richard, Xiaoyin Chen , Moksh Jain , Sungjin Ahn , Sungsoo Ahn, Yoshua Bengio

I’m not just dreaming of an AGI that answers hard questions. I’m dreaming of one that can create entirely new theories — like how Claude Shannon created information theory.

Excited to present Monte Carlo Tree Diffusion (MCTD) at ICML Conference ICML2025! We integrate diffusion models with Monte Carlo Tree Search for more scalable planning. Come see how MCTD consistently outperforms other methods in complex, long-horizon tasks. 🗓️ Wednesday (Day 2) ⏰