Bhavya Sukhija
@sukhijabhavy
PhD Student at @eth_en, Research visitor @berkeley_ai, Research Scientist intern @AWS. Artificial Intelligence, Robotics, Reinforcement Learning.
Introducing CoordTok, a scalable video tokenizer that can encode a 128-frame video into only 1k tokens. CoordTok learns a mapping from (x, y, t) coordinates to the corresponding patches of input videos. 🧵[1/6] project page: huiwon-jang.github.io/coordtok/
🐕 I'm happy to share my paper: RAMBO: RL-augmented Model-based Whole-body Control for Loco-manipulation has been accepted by IEEE Robotics and Automation Letters (RA-L) 🧶 Project website: jin-cheng.me/rambo.github.i… Paper: arxiv.org/abs/2504.06662
EXAIT workshop is completely full due to Sergey Levine’s talk 🤣 Come to our poster at 11:45-2:15!! exait-workshop.github.io/schedule/
Clinical notes are messy, inconsistent, and unstructured—yet they hold some of the most valuable signals in real-world clinical practice. Join us today at ICML at the Foundation Models for Structured Data workshop to see how we can make sense of these notes! 📍 West Ballroom D
Sim2real is getting so mature that with great hardware (thanks @clemens_chr @katzschmann), you can get things running in just days 🤖 We are also demoing @orcahand trained with MuJoCo Playground at #RSS2025. Come by Associates Park at 12:30pm-2pm and 6:30pm-8pm! @RoboticsSciSys
📢 Deadline extension! Submit your work to EXAIT at #icml2025 — now accepting papers until May 31st AoE.
What is the place of exploration in today's AI landscape and in which settings can exploration algorithms address current open challenges? Join us to discuss this at our exciting workshop at @icmlconf 2025: EXAIT! exait-workshop.github.io #ICML2025
Looking forward to presenting my work on MaxInfoRL on Saturday, 26.04 5:30 pm at @iclr_conf in Singapore. Will also present my work on MBRL (openreview.net/pdf?id=VGdqa79…) at the Robot Learning workshop on 27.04! Happy to chat about these works or RL and robotics in general!
🚨 New reinforcement learning algorithms 🚨 Excited to announce MaxInfoRL, a class of model-free RL algorithms that solves complex continuous control tasks (including vision-based!) by steering exploration towards informative transitions. Details in the thread 👇
What is the place of exploration in today's AI landscape and in which settings can exploration algorithms address current open challenges? Join us to discuss this at our exciting workshop at @icmlconf 2025: EXAIT! exait-workshop.github.io #ICML2025
We've just added the @boosterobotics T1 Humanoid to MuJoCo Playground, with code to train a joystick controller as well as a deployable ONNX file. A super fun weekend with @carlo_sferrazza and @qiayuanliao :) github.com/google-deepmin…
Very excited to share that my paper together with @sukhijabhavy about Active Exploration on Constrained Markov decision processes got accepted to ICLR 2025 🥳🥳🥳🎉🎉🎉🎉🎉🎉
The ultimate test of any physics simulator is its ability to deliver real-world results. With MuJoCo Playground, we’ve combined the very best: MuJoCo’s rich and thriving ecosystem, massively parallel GPU-accelerated simulation, and real-world results across a diverse range of…
Ever wondered what robots 🤖 could achieve if they could not just see – but also feel and hear? Introducing FuSe: a recipe for finetuning large vision-language-action (VLA) models with heterogeneous sensory data, such as vision, touch, sound, and more. Details in the thread 👇
🎉 My Master thesis, done under the supervision of @pkassraie_ and @arkrause, is finally online! I worked on contextual bandit guarantees for fine-tuning. The theory-practice gap is wide, but we argue that pre-training can help narrowing it. arxiv.org/abs/2501.06258 A short 🧵
Papers Alert 🚨: Three papers from @crl_ethz got accepted to @NeurIPSConf! Details below! 🧶 @ETH @ETH_en #NeurIPS24