🔥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

The proposed presidential budget cuts available non-military basic science funding in the US by over half, probably the largest cut ever. It is not compensated by additional funding elsewhere.

Totally wasted (leveraged?!) the entire morning learning about the state of the art of Minecraft AI agents. If you are, like me, prone to ratholing on topics like this, you will want to avoid the YouTube channel I'm about to post in the comments.

Not sure why the gutting of American science funding isn’t a bigger story. No one voted for it, it reduces American innovation and economic competitiveness in the near-term and long-term, and it isn’t even being done efficiently, if that were in fact the goal.

Totally agree. I’ve always felt incredibly lucky to have Joyce as my advisor, not just because she helped me turn some of my wildest research ideas into real experiments and solid papers, but because she showed me what it means to be a researcher beyond the papers. She’s been a

What's with these arguments over whether X or Y or whatever was the first LLM RL library? These all came in the last 3 months We wrote multiturn RL4LMs like 3+ years ago github.com/allenai/RL4LMs There were other simple versions even before. ML ppl approaching goldfish memory

People should stop arguing over which RL x LLMs framework came first and start arguing over which one scales best. That will determine long-term success

Mindcraft now has an official research paper behind it! We focused on the collaboration AI agents in minecraft, and introduce mindcraft as a framework for future research. I made a video to showcase it! Thanks to coauthors Kolby Nottingham Isadora White Prithviraj (Raj) Ammanabrolu Lianhui Qin Paper,

Now accepted at ICML 2025

We didn't make too big a deal of it but this is what we did for our multi agent MINDcraft env github.com/kolbytn/mindcr… Policy latency is an issue with real time embodied tasks so we have ways of summarizing& reusing reasoning tokens. Lots of cool eng behind the scenes

Are you an RL PhD at Harvard who has had your funding wrecked by the government and working on topics related to multi-agent? Reach out, I am happy to try to find a way to support you

"Foundation" models for embodied agents are all the rage but how to actually do complex looong context reasoning? Can we scale Beyond Needle(s) in the (Embodied) Haystack? ∞-THOR is an infinite len sim framework + guide on (new) architectures/training methods for VLA models

1) What

Goal-conditioned RL (GCRL) is great - unsupervised, can use data (in offline mode), flexibility to define tasks at test time. But can we run GCRL on *language data*?? In our new work we show that language GCRL enables sophisticated test-time reasoning for interactive tasks! 🧵👇

ACADEMICS: it is time to get our heads out of our *sses. This is not the moment for personal ambition, why your latest sophisticated widget beats rivals intricate theorem. The scientific franchise is under attack. It is time to defend it to the public. x.com/davidbau/statu…

🤝 Humans + AI = Better together? Our #ACL2025 tutorial offers an interdisciplinary overview of human-AI collaboration to explore its goals, evaluation, and societal impacts 🤖

#ICML #cognition #GrowAI We spent 2 years carefully curated every single experiment (i.e. object permanence, A-not-B task, visual cliff task) in this dataset (total: 1503 classic experiments spanning 12 core cognitive concepts). We spent another year to get 230 MLLMs evaluated