🔥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

🎉 Excited to share that "NullModel" has been accepted to #ICLR2025 as an oral presentation! I am on the Job Market! I am seeking a Research Scientist position aligned with, but not limited to Data-Centric AI and AI Safety. Feel free to reach out if interested. RT appreciated!

Current robot learning methods are good at imitating tasks seen during training, but struggle to compose behaviors in new ways. When training imitation policies, we found something surprising—using temporally-aligned task representations enabled compositional generalization. 1/

How different are the outputs of various LLMs, and in what ways do they differ? Turns out, very very different, up to the point that a text encoding classifier could tell the source LLM with 97% accuracy. This is classifying text generated by LLMs, between ChatGPT, Claude,

A lot of work focuses on test-time scaling. But we aren't scaling it optimally, simply training a long CoT doesn't mean we use it well. My students developed "v0" of a paradigm to do this optimally by running RL with dense rewards = minimizing regret over long CoT episodes. 🧵⬇️

Papers accepted at ICLR 2026 2025: - An Effective Theory of Bias Amplification arxiv.org/abs/2410.17263 - Pitfalls of Memorization arxiv.org/abs/2412.07684 - Strong Model Collapse arxiv.org/abs/2410.04840 - Beyond Model Collapse arxiv.org/abs/2406.07515 With Julia Kempe,

We refused to cite the paper due to severe misconduct of the authors of that paper: plagiarism of our own prior work, predominantly AI-generated content (ya, the authors plugged our paper into an LLM and generated another paper), IRB violations, etc. Revealed during a long

I need to offer some clarification for this post because it would be wrong for people to lump this situation in with ones where work that is considered low quality (or inconvenient for priority arguments) is not cited. (This is my gut reaction whenever citations are excluded.)

Check out our poster tmr at 10am at the ICLR Bidirectional Human-AI Alignment workshop! We cover how on-policy preference sampling can be biased and our optimal response sampling for human labeling. NYU Center for Data Science AI at Meta Julia Kempe Yaqi Duan x.com/feeelix_feng/s…

🚨 Your RL only improves 𝗽𝗮𝘀𝘀@𝟭, not 𝗽𝗮𝘀𝘀@𝗸? 🚨 That’s not a bug — it’s a 𝗳𝗲𝗮𝘁𝘂𝗿𝗲 𝗼𝗳 𝘁𝗵𝗲 𝗼𝗯𝗷𝗲𝗰𝘁𝗶𝘃𝗲 you’re optimizing. You get what you optimize for. If you want better pass@k, you need to optimize for pass@k at training time. 🧵 How?