🔥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

If you're at NeurIPS, come tomorrow for the Oral+Poster on "Learning Formal Mathematics from Intrinsic Motivation"! Really fun work with David Broman Nick Haber noahdgoodman that put together much of what I did in the past years, w/ a new twist with open-ended learning!

It was really fun to have been a part of this -- so excited to see it out there!

One of the biggest problems with VLMs today is their over-reliance on language semantics. We introduce Symmetrical Visual Contrastive Optimization (S-VCO), a simple posttraining method for enhancing VLMs' ability to discern visual details.

Some exciting new work of ours led by Shengguang Wu with Fan-Yun Sun and Kaiyue Wen!

Super excited to get this out there! Work led by Fan-Yun Sun and Weiyu Liu, with Siyi Gu, @dill_pkl, Goutam Bhat, Federico Tombari, Manling Li, and Jiajun Wu Project site: ai.stanford.edu/~sunfanyun/lay…

At ICLR! Say hi.

Go see Logan Cross present!

Working with Sergey Pankov on an unbiased estimator of the gradient for attention, that has very low variance even considering in expectation a small number of backwards tokens.

1/ I'm excited to share recent results from my first collaboration with the amazing Aran Nayebi and Leo Kozachkov! We show how autonomous behavior and whole-brain dynamics emerge in embodied agents with intrinsic motivation driven by world models.

As we optimize model reasoning over verifiable objectives, how does this affect human understanding of said reasoning to achieve superior collaborative outcomes? In our new preprint, we investigate human-centric model reasoning for knowledge transfer 🧵: