Rajat Kumar Jenamani

Really excited to share that FEAST won the Best Paper Award at #RSS2025! Huge thanks to everyone who’s shaped this work, from roboticists to care recipients, caregivers, and occupational therapists. ❤️

A wonderful project. The three takeaways are that home robots need adaptability, safety, and transparency to be deployed around people; if a machine is to be doing something as intimate as sharing your personal space at home and helping you eat, you need to be able to trust it,…

Are current eval/deployment practices enough for today’s robot policies? Announcing the Eval&Deploy workshop at CoRL 2025 @corl_conf, where we'll explore eval + deployment in the robot learning lifecycle and how to improve it! eval-deploy.github.io 🗓️ Submissions due Aug 30

Meet Emma - our AI scout! She scans EVERY vine for disease so that farmers maximize yield and produce better wine. 🍷 Emma's scouting 10 acres for free! DM or go to budbreak.tech/contact to have Emma scout your fields.

Using OT to define rewards for imitating video demos is popular, but it breaks down when demos are temporally misaligned—a frequent challenge in practice. We present ORCA at #ICML2025 , which defines rewards by aligning sequences, rather than matching individual frames via OT.

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 👇

How can 🤖 learn from human workers to provably reduce their workload in factories? Our latest @RoboticsSciSys paper answers this question by proposing the first cost-optimal interactive learning (COIL) algorithm for multi-task collaboration.

VLAs have the potential to generalize over scenes and tasks, but require a ton of data to learn robust policies. We introduce OG-VLA, a novel architecture and learning framework that combines the generalization strengths of VLAs with the robustness of 3D-aware policies. 🧵

Current robot policies often face a tradeoff: they're either precise (but brittle) or generalizable (but imprecise). We present ViTaL, a framework that lets robots generalize precise, contact-rich manipulation skills across unseen environments with millimeter-level precision. 🧵

@TapoBhat is doing very important work in real-world HRI and could not be more deserving of this award!

Teleoperation is slow, expensive, and difficult to scale. So how can we train our robots instead? Introducing X-Sim: a real-to-sim-to-real framework that trains image-based policies 1) learned entirely in simulation 2) using rewards from human videos. portal-cornell.github.io/X-Sim

Just a small reminder that our workshop is happening tomorrow, and we have an amazing line of speakers! Make sure to check out the workshop website for the schedule. 🤖

At #RSS2025! I’m presenting FEAST on Mon, June 23: spotlight talk in the morning HRI session, poster in the afternoon. Happy to chat about in-home robot deployment, personalization, and my recent interest in enforcing constraints on end-to-end policies for critical tasks.

Absolutely beautiful. I love research applying robotics to accessibility.

This will be a game-changer for parents, imagine not needing to stand there trying to feed your child while they watch cocomelon or youtube shorts

Such a powerful use of robotics. Love seeing this new paper take real steps toward making assistive feeding robots a reality.

Robotics research often takes time to translate into real-world impact, but every so often, a project comes along that delivers immediate and tangible benefits to the society.

Congrats @rkjenamani and the entire team @EmpriseLab on this impressive accomplishment and being nominated for Best Paper Award and Best Systems Paper Award at #RSS 2025! This project took almost 2.5 years to get to this stage, and I am incredibly proud of what we have achieved…