Baifeng

@baifeng_shi

@physical_int, @berkeley_ai, vision, robotics; ex @nvidia; opinions are my own.

bfshi.github.io

Joined September 2021

Replies Reposts Media

Pinned
Baifeng
@baifeng_shi
Mar 24
Humans can see in high-res, high-FPS in real-time. Why can't VLMs? Introducing AutoGaze: ViTs/VLMs "gaze" only at key video regions! Up to 4-100x token savings, 19x speedup, and enables scaling to 4K-res 1K-frame videos. 📄 arxiv.org/abs/2603.12254 🌐 autogaze.github.io 🤗
00:00
47
Baifeng
@baifeng_shi
Jun 20
Learning from task-agnostic, explorative experience!
Junyi Zhang
@junyi42
Jun 19
Children learn from play. Can robots do the same? We propose 𝐏𝐥𝐚𝐲𝐟𝐮𝐥 𝐀𝐠𝐞𝐧𝐭𝐢𝐜 𝐑𝐨𝐛𝐨𝐭 𝐋𝐞𝐚𝐫𝐧𝐢𝐧𝐠, a paradigm that gives embodied coding agents a play stage before downstream tasks arrive, and instantiate it with 𝐑𝐀𝐓𝐬 (Robotics Agent Teams), where robots
00:00
Baifeng
@baifeng_shi
Apr 29
Congrats on the release!
Hanrong YE @ Nvidia
@leoyerrrr
Apr 28
@nvidia introduces Nemotron 3 Nano Omni LLMs. A year ago, we started research on omni-modal LLMs—covering everything from architecture to data—and released OmniVinci (ICLR 2026). We’ve received some feedback since then, but two clear themes emerged: 9B parameters weren't quite
Baifeng
@baifeng_shi
Apr 17
Our newest generalist model with great zero-shot capabilities! Some highlights: - can utilize both high-quality and low-quality data via proper conditioning - careful design of diverse data - can also condition on imagined goal image from an image world model - as a result, the
Physical Intelligence
@physical_int
Apr 16
Our newest model, π0.7, has some interesting emergent capabilities: it can control a new robot to fold shirts for which we had no shirt folding data, figure out how to use an appliance with language-based coaching, and perform a wide range of dexterous tasks all in one model!
00:00
2
Baifeng
@baifeng_shi
Apr 9
Check out @LongTonyLian’s cool parallel reasoning work!
Long Lian
@LongTonyLian
Apr 8
Our parallel reasoning project ThreadWeaver is now open-sourced 🎉! Check out our Data Gen/SFT/RL recipe at github.com/facebookresear… In case you don't know, ThreadWeaver 🧵⚡️ is the first parallel reasoning method to achieve comparable reasoning performance to widely-used