🔥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

Please come to our poster today — I’ll be there to present our work! “A Coefficient Makes SVRG Effective” Friday, 3:00–5:30 Hall 3 + Hall 2B #385 iclr.cc/virtual/2025/p…

Not at ICLR myself, but David Yin will kindly give the oral and poster presentation on my behalf, for our dataset bias paper. David knows the work as well as I know it (if not better). Check it out, happening today! Oral: Saturday, 10:42 — 10:54am, Peridot 202-203

Watching Zhuang Liu's - "Transformers without Normalization", this slide is a reminder how our optimizer and architecture choices are coupled

This is exactly like how students in competitive math do. Thinking really hard on one problem and trying to find different solutions for one problem It is still quite wild… the model does not overfit to that one example

🤣it happens when you can’t find the emergency stop

Accepted to #ICML 25 & also recently featured in CMU news and Fast Company: cs.cmu.edu/news/2025/llm-… fastcompany.com/91286162/ai-ch…

The post-training dataset is fully open!

🧵1/ Everyone says toxic data = bad models. But what if more toxic data could help us build less toxic models? Our new paper explores this paradox. Here’s what we found 👇

I had the same experience with ChatGPT. Once it misunderstands a question or gets something wrong, it will never go back to normal with more clarification. Pro tip: open a new tab each time

Please come to our oral and poster presentations on Wednesday — I’ll be there along with Zekai Wang and Dantong Niu to present our work! Oral: Wednesday, 3:20 — 3:25pm, Room 314 Poster: Wednesday, 3:50 — 4:25pm, Room 314 Paper: arxiv.org/abs/2410.12782

What an exciting #ICRA2025 week!✨ Catch our Oral & Poster this Wednesday in the Robotic Foundation Models Session — with David Yin, Zekai Wang and Dantong Niu Oral: Wed, 3:20 — 3:25pm, Room 314 Poster: Wed, 3:50 — 4:25pm, Room 314 Project: davidyyd.github.io/roboprompt/

While we're at "recognizing evals", here is a legendary vision paper from CVPR 2011 It shows that it's quite easy to classify which dataset an image comes from (39% acc, vs random=8%). The point being, every dataset having its distinct signature should be a default assumption.

Check out our new paper “Generative Modeling of Weights: Generalization or Memorization?” — we find that current diffusion-based neural network weight generators often memorize training checkpoints rather than learning a truly generalizable weight distribution!