Pratap Tokekar retweeted
I'm excited to be at IROS 2026 in Pittsburgh, America's Steel City! I'll be presenting and participating in a panel at the Workshop on Foundation Models for Multi-Robot Systems. If this work intrigues you, see @ptokekar's post on postdoc openings at Amazon Robotics Fleet Sciences
Our Fleet Science team at Amazon Robotics will also have a strong presence at IROS. We have multiple open postdoc positions.. come say hi! xitter.cf/ptokekar/status/206557…
Pratap Tokekar retweeted
We’re hiring a postdoc at NTU Singapore to push the frontiers of robot planning and learning.
Further details about the application process can be found here: algrobogroup.github.io/oppor…
Thank you for your support in sharing this opportunity!
#Hiring #Postdoc #Robotics #NTUsg
The Fleet Science team at Amazon Robotics has two open postdoctoral positions. I've been working with the exceptional folks in the Fleet Science team. This is a unique opportunity to do cutting-edge science and have a real-world impact at Amazon scale.
(2) Multi-Agent Path Planning & Explainable AI
- Multi-agent path planning using learning and optimal transport
- Explainability and interpretability techniques for DeepFleet Foundation Models
amazon.jobs/en/jobs/3117815/…
Two new papers at #CVPR2026 and #AAMAS2026 from @ruiliu0 and Chak Lam Shek from my lab. Rui and Chak are both on the job market this year!
Stable and Efficient Single-Rollout RL for Multimodal Reasoning
@ruiliu0 et al. #CVPR2026
Paper: arxiv.org/pdf/2512.18215
Multi-Agent Trust Region Policy Optimisation: A Joint Constraint Approach
Chak Lam Shek, @guangyao_shi, and Pratap Tokekar
Paper: arxiv.org/pdf/2508.10340 #AAMAS2026
Pratap Tokekar retweeted
I am heading to #CVPR2026 next week to present our work MSSR (Multimodal Stabilized Single-Rollout). This work aims to improve the training efficiency of multimodal RLVR by using just single rollout instead of group-based approaches such as GRPO.
With around 1/4 of training steps, our approach can reach similar accuracy as GRPO. While trained for the same amount of compute and test on multiple benchmarks for generalization performance, our approach can outperform other group-based baselines.
Thanks to all my amazing collaborators Dian Yu, Lei Ke, @HaolinLiu616, @YujunZhou0017, @LiangZhenwen, @haitaominlp, @ptokekar, Dong Yu.
I will also be actively looking for full time opportunities, very happy to connect!
Paper: arxiv.org/pdf/2512.18215
Project page: mssr-proj.github.io
Congratulations Dr. @peihong_yu Proud to be hooding and celebrating your well deserved success!
Pratap Tokekar retweeted
Join us and @JohnsHopkinsIAA on Monday for a joint talk on trends in safe and reliable reinforcement learning, featuring @JHUAPL’s @aekoppel and @umdcs & @amazon Robotics’ @ptokekar. Learn more here: cs.jhu.edu/event/iaa-cs-semi…
ALT Institute for Assured Autonomy & Computer Science Seminar Series. Talk Trends in Safe and Reliable Reinforcement Learning. February 9, 2026, 1–2 p.m. Zoom. Alec Koppel, Johns Hopkins APL. Pratap Tokekar, UMD & Amazon.
Pratap Tokekar retweeted
During my phd work (fish-finding robots @UMNCSE ), our record low during a field op was -15. It is particularly difficult to type when the laptop (lcd!) screen refresh is 1hz, and your fingers are numb claws.
Wouldn't trade that memory for anything.
Pratap Tokekar retweeted
Interested in theoretical RL?
Find us at poster #3312 at #NeurIPS2025
🗓 Thu Dec 4, 4:30–7:30 PM
📍 Exhibit Hall C/D/E
@anas__barakat @SOURADIPCHAKR18 @ptokekar @peihong_yu
Excited to be at #NeurIPS2025 this week. Would love to meet folks interested in robotics. We will be presenting two papers...
CAML: Collaborative Auxiliary Modality Learning for Multi-Agent Systems
Thu 4 Dec 4:30 p.m. PST — 7:30 p.m. openreview.net/forum?id=OhUu…
@ruiliu0 @gammaumd
Would especially love to chat with students interested in doing a PhD @umdcs and super cool work that we're doing at Amazon Robotics