Utkarsh Mishra

How can robots compositionally generalize over multi-object multi-robot tasks for long-horizon planning? At #CoRL2024, we introduce Generative Factor Chaining (GFC), a diffusion-based approach that composes spatial-temporal factors into long-horizon skill plans. (1/7)

Large robot datasets are crucial for training 🤖foundation models. Yet, we lack systematic understanding of what data matters. Introducing MimicLabs ✅System to generate large synthetic robot 🦾 datasets ✅Data-composition study 🗄️ on how to collect and use large datasets 🧵1/

If you are interested in solving complex long-horizon tasks, please join us at the 3rd workshop on Learning Effective Abstractions for Planning (LEAP) at @corl_conf! 📅 Submission deadline: Sep 5 🐣 Early bird deadline: Aug 12

We're excited to announce the third workshop on LEAP: Learning Effective Abstractions for Planning, to be held at #CoRL2025 @corl_conf! Early submission deadline: Aug 12 Late submission deadline: Sep 5 Website link below 👇

Tired of slow-moving robots? Want to know how learning-driven robots can move closer to industrial speeds in the real world? Introducing SAIL - a system for speeding up the execution of imitation learning policies up to 3.2x on real robots. A short thread: 1/

Despite advances in end-to-end policies, robots powered by these systems operate far below industrial speeds. What will it take to get e2e policies running at speeds that would be productive in a factory? It turns out simply speeding up NN inference isn't enough. This requires…

How annealing helps overcoming multimodality? In our ICLR 2025 paper openreview.net/forum?id=P6IVI… and preprint arxiv.org/abs/2502.04575, we established the first complexity bound for annealed sampling and normalizing constant (⇔free energy) estimation under weak assumptions on target!

Introducing EgoMimic - just wear a pair of Project Aria @meta_aria smart glasses 👓 to scale up your imitation learning datasets! Check out what our robot can do. A thread below👇

Evaluation in robot learning papers, or, please stop using only success rate a paper and a 🧵 arxiv.org/abs/2409.09491

DP-Attacker is accepted to #NeurIPS2024 🎉

#ICLR2024 **The LDM is Vulnerable because of the Encoder! ** Our paper: Towards More Effective Protection Against Diffusion-Based Mimicry with Score Distillation has been accepted to ICLR 2024! 🌟 github.com/xavihart/Diff-…

Since we are entering the "BC is all you need" phase of Robot Learning😜 --- Robomimic (robomimic.github.io) allows you to play with SOTA algorithms (BC-Transformer, DiffusionPolicy, etc.) on challenging tasks. Also easy to integration with physical robots!

Delighted to announce the successful defense of my exceptional PhD student @qsh_zh @GTrobotics @mlatgt @gatechengineers So proud of the remarkable achievements he has made, especially in diffusion models. It has been a privilege to mentor such a brilliant mind #ProudAdvisor

There was a lot of good and interesting debate on "is scaling all we need to solve robotics?" at #CoRL23. I spent some time writing up a blog post about all the points I heard on both sides: nishanthjkumar.com/Will-Scaling-S…