Humans use pointing to communicate plans intuitively. Compared to language, pointing gives more precise guidance to robot behaviors. Can we teach a robot how to point like humans? Introducing RoboPoint 🤖👉, an open-source VLM instruction-tuned to point. Check out our new work:

Happy to share a new preprint: "Coloring Between the Lines: Personalization in the Null Space of Planning Constraints" w/ Rajat Kumar Jenamani, Ziang Liu, Ben Dodson, and Tapomayukh "Tapo" Bhattacharjee. TLDR: We propose a method for continual, flexible, active, and safe robot personalization. Links 👇

Excited to share our work on continual, flexible, active, and safe robot personalization w/ Tom Silver, Ziang Liu, Ben Dodson & Tapomayukh "Tapo" Bhattacharjee. Also: Tom Silver is starting a lab at Princeton!! I HIGHLY recommend joining — thoughtful, kind, and an absolute joy to work with!

🔥Best Paper Award at #ICRA2025 Thrilled to share that our paper MAC-VO has been awarded the 𝘽𝙚𝙨𝙩 𝘾𝙤𝙣𝙛𝙚𝙧𝙚𝙣𝙘𝙚 𝙋𝙖𝙥𝙚𝙧 𝘼𝙬𝙖𝙧𝙙 and the 𝘽𝙚𝙨𝙩 𝙋𝙖𝙥𝙚𝙧 𝘼𝙬𝙖𝙧𝙙 𝙤𝙣 𝙍𝙤𝙗𝙤𝙩 𝙋𝙚𝙧𝙘𝙚𝙥𝙩𝙞𝙤𝙣! Check our project: mac-vo.github.io

System ID for legged robots is hard: (1) Discontinuous dynamics and (2) many parameters to identify and hard to "excite" them. SPI-Active is a general tool for legged robot system ID. Key ideas: (1) massively parallel sampling-based optimization, (2) structured parameter space,

Claude Sonnet 4 on ARC-AGI Semi Private Eval Base * ARC-AGI-1: 23%, $0.08/task * ARC-AGI-2: 1.2%, $0.12/task Thinking 16K * ARC-AGI-1: 40%, $0.36/task * ARC-AGI-2: 5.9%, $0.48/task Sonnet 4 sets new SOTA (5.9%) on ARC-AGI-2

🚨Excited to announce the 1st Workshop on Vision Meets Physics at @CVPR2025! Join us on June 12 for a full-day event exploring the synergy between physical simulation & computer vision to bridge the gap between the virtual and physical worlds. URL: tinyurl.com/vis-phys

Zero fluff, maximum insight ✨. Let’s see what LLMs are really made of, with 🧠 Brainteasers. We’re not grading answers 🔢. We’re grading thinking 💭. Brute force? Creative leap? False confession? 🤔 Instead of asking “Did the model get the right answer?”, we ask: “Did it

Introducing LocoTouch: Quadrupedal robots equipped with tactile sensing can now transport unsecured objects — no mounts, no straps. The tactile policy transfers zero-shot from sim to real. Core Task-Agnostic Features: 1. High-fidelity contact simulation for distributed tactile

Excited to share work on using classical search approaches to scale inference in diffusion models! We show how global graph search algorithms (BFS, DFS) and local search can be used to improve generation performance across domains such as image generation, planning, and RL!

Tired of watching fancy humanoid dancing? Can they just do some daily useful tasks like: "Pass me a bottle of Water🍺"? 🤔Turns out it's nontrivial to stablize whole-body manipulation and locomotion at the same time. We basically want our humanoid to be stable as a camera

Bored of linear recurrent memories (e.g., linear attention) and want a scalable, nonlinear alternative? Our new paper “Test-Time Training Done Right” propose LaCT (Large Chunk Test-Time Training) — a highly efficient, massively scalable nonlinear memory with: 💡 Pure PyTorch

Transformer-based neural networks achieve impressive performance on coding, math & reasoning tasks that require keeping track of variables and their values. But how can they do that without explicit memory? 📄 Our new ICML paper investigates this in a synthetic setting! 🧵 1/13

Are world models necessary to achieve human-level agents, or is there a model-free short-cut? Our new #ICML2025 paper tackles this question from first principles, and finds a surprising answer, agents _are_ world models… 🧵

Introducing Mobi-π: Mobilizing Your Robot Learning Policy. Our method: ✈️ enables flexible mobile skill chaining 🪶 without requiring additional policy training data 🏠 while scaling to unseen scenes 🧵↓

Is RL really scalable like other objectives? We found that just scaling up data and compute is *not* enough to enable RL to solve complex tasks. The culprit is the horizon. Paper: arxiv.org/abs/2506.04168 Thread ↓

Check out this MIT News article about our research on GPU-accelerated manipulation planning! news.mit.edu/2025/new-syste… William Shen Nishanth Kumar Ankit Goyal

Real-world RL, where robots learn directly from physical interactions, is extremely challenging — especially for high-DoF systems like mobile manipulators. 1⃣ Long-horizon tasks and large action spaces lead to difficult policy optimization. 2⃣ Real-world exploration with

Introducing UFM, a Unified Flow & Matching model, which for the first time shows that the unification of optical flow and image matching tasks is mutually beneficial and achieves SOTA. Check out UFM’s matching in action below! 👇 🌐 Website: uniflowmatch.github.io 🧵👇

This week's #PaperILike is "Long-Horizon Multi-Robot Rearrangement Planning for Construction Assembly" (Hartmann et al., TRO 2022). Take two minutes to watch this video: youtube.com/watch?v=Gqhouv… I don't use a lot of emojis, but 🤯 PDF: arxiv.org/abs/2106.02489