How can we leverage VLAs to learn to solve complex tasks that fall outside their typical capabilities? We introduce Semantic Action RL—treat the VLA’s language prompt as an action and optimize this with RL! Check out Jagdeep's thread for all the details! semantic-action-rl.github.io

If you want a robot to do something well, you need to know how to talk to it. If you don't, you can learn, with Semantic Action RL! In our paper, @JagdeepBhatia8, @ajwagenmaker, @verityw_ show how RL over VLA prompts enables new tasks and learns blazing fast in the real world!

How can we elicit useful, semantically meaningful behaviors from generalist policies? We introduce Flow Reversal Steering (FRS) as a method to refine coarse, semantically meaningful commands into effective robot actions! flow-reversal-steering.githu… 1/N

Flow reversal steering allows "steering" diffusion-based VLAs with high-level actions, for example from VLM reasoning. This also lets us run RL in the diffusion noise space with exploration guided by high-level reasoning: think through a task, then practice it! 👇

🤔 Can we train one VLA policy to control multi-robot teams without any explicit communication? ✨ Introducing CHORUS: a single policy for decentralized, multi-embodiment collaboration 🧵⬇️

Excited to be in Rio attending ICLR this week to present some papers! 🧵(1/3) (1) Decoupled Q-Chunking w/ @seohong_park and @svlevine on Fri 3:15-5:45 (#4504) x.com/qiyang_li/status/19996…

A central challenge in #physical #AI is data scarcity: vision-language-action (#VLA) models are fundamentally limited by the availability of high-quality robotics demonstrations. In our recent work, we introduce R&B-EnCoRe (arxiv.org/pdf/2602.08167), a framework that enables models to self-bootstrap embodied #reasoning by leveraging synthetic visuo-textual data together with limited embodiment-specific experience. In essence, R&B-EnCoRe allows models to learn how to reason in an embodied setting. Our approach treats reasoning as a latent variable and uses self-supervised refinement to learn reasoning strategies that are directly predictive of successful control—without human annotations, reward engineering, or external verifiers. We validate the approach across a range of embodiments—including manipulation, navigation, and autonomous driving—and across model scales from 1B to 30B parameters, observing consistent improvements: 💪 28% task success in real-world manipulation 🦿 101% score in legged locomotion navigation 🚗 −21% collision rate in autonomous driving Overall, this work highlights a promising direction: aligning internet-scale priors with embodiment-specific data to enable scalable, self-improving physical intelligence. Kudos to an amazing team: Milan Ganai Katie Luo @JonasFrey96 Clark Barrett 🌐 Website: milanganai.github.io/rnb-enc… 📄 Paper: arxiv.org/pdf/2602.08167

A reward model that works, zero-shot, across robots, tasks, and scenes? Introducing Robometer: Scaling general-purpose robotic reward models with 1M trajectories. Enables zero-shot: online/offline/model-based RL, data retrieval IL, automatic failure detection, and more! 🧵 (1/12)

This one has been a long time coming: today we’re introducing MEM, an approach for giving VLAs short-term and long-term memory. Memory is such an obvious capability, but adding it isn’t easy (most VLAs today are memory-less). A short thread on challenges, solutions, and the new capabilities MEM unlocks for us.

If we train VLAs to respond to diverse multimodal prompts, then we can steer them better: [grasp the carrot]/[move to x,y,z]/[put the carrot on the plate]. With many levels of detail, powerful VLMs can step in and steer the model to success much more often! More below 👇