#ObjectReact and its predecessor #RoboHop were inspired by Grey Walter's "tortoise" #robots, after reading about #reactive #navigation in Peter Corke's book. (Though our approach is only locally reactive, given a global plan) #CoRL2025 share.google/bH9PXniOOZwtIM…

Everyone posting about the Dwarkesh interview (including Dwarkesh himself!) is missing this subtle point. When LLMs imitate, they imitate the ACTION (ie the token prediction to produce the sequence). When humans imitate, they imitate the OUTPUT but must discover the action

Very insightful Keynote by Wenzen Yuan at Conference on Robot Learning on #tactile sensing: - variance in sensor design a bigger issue than the variance in what's sensed by a single sensor - you can simulate #touch with #vision #CoRL2025

Sangbae Kim's (Sangbae Kim) keynote talk Conference on Robot Learning had a philosophical take on #imitation, #thinking and #learning, inspired by Oscar Wilde "The imagination imitates. It is the critical spirit that creates ..." #CoRL2025 #robotics #AI

It just dawned on me that this is tiktok for intellectuals-ish. We also have sudden trends where everybody does their version of the trends. Just in the last few days: - Thinky blogpost is not news - bitter lesson about Rich - Frontier secret vs GRPO It's fun though, ngl

We are presenting this now for next 2 hours, drop by at the corner of the exhibition hall! Conference on Robot Learning #CoRL2025 Dustin Craggs Vineeth Bhat Feras Dayoub

One of the cute ones #CoRL2025 Chatting with Feras Dayoub

K-Pop performance Conference on Robot Learning banquet by Geenius #CoRL2025

Computer Vision "folk" spotted at #CoRL2025 (as usual) Conference on Robot Learning #CVPR2026

Heya roboticists & brain engineers ~ Here's a list of 200+ robotics papers from Conference on Robot Learning -- with links to papers, TLDRs, project pages & code So much incredible research this year. Congrats to all the authors moving our field forward 🫡 github.com/smallfryy/corl…

Q: Pick something which is neither green nor red, neither cube nor cuboid, and place it on something, which is neither white nor red A: blue cylinder in blue tray (Also failed once) Google DeepMind #CoRL2025

What an incredible night! Thank you Conference on Robot Learning for shaking things up and inviting both a K-pop band and a Korean traditional music band to the CoRL Banquet. It made the evening truly special.

This was another important slide from Sangbae Kim, comparing #reinforcement learning with #MPC (Model Predictive Control) #CoRL2025

This year my #ICCV2025 AC job was much easier because I had already sampled from this future distribution😁 Some reviewers that I rated "exceeds expectations" didn't make it to this list, it might be due to reviewing fewer or withdrawn papers, not sure (latter's less likely tho)

A reminder that Conference on Robot Learning talks are fully and freely available to watch on YouTube Keynotes are a must watch! Conference on Robot Learning youtube.com/@conferenceonr… #CoRL2025

Introducing Kaleido💮 from AI at Meta — a universal generative neural rendering engine for photorealistic, unified object and scene view synthesis. Kaleido is built on a simple but powerful design philosophy: 3D perception is a form of visual common sense. Following this idea,

Our #NeurIPS2025 *spotlight* paper #SegMASt3R establishes image segment matching as a benchmark task & enables high performance downstream on 3D Instance Mapping & Object-Relative Navigation segmast3r.github.io Huge effort by Rohit Jayanti Swayam Agrawal Vansh Robotics Lab

Finally someone citing proper sources! Thanks Rohit Jayanti Sourav Garg The paper is a NeurIPS spotlight btw

Chris Offner Alexandre Morgand That's a curve ball question but here's my intuition/hypothesis: 1. Locality of Task: Depth estimation driven by priors can be primarily thought of as a local task, i.e., given semantic context of things in the image you can predict the relative depth - hence why also linear