Snehal Jauhri

Snehal Jauhri

I'm a Post-Doctoral researcher at the 3D AI Lab at the Technical University of Munich, working with Angela Dai. My research focuses on 3D visual scene understanding and generation, reinforcement learning, and embodied AI.
I completed my PhD in Robot Perception and Learning at the PEARL Lab at TU Darmstadt, supervised by Georgia Chalvatzaki.
Previously, I was a PhD Research Scientist Intern at the Allen Institute for AI, working on large-scale simulation and vision-language models MolmoSpaces and MolmoB0T.
I obtained my Master's degree from TU Delft, working with Jens Kober on machine learning for robotics.

Email / CV / Scholar / LinkedIn / Twitter / GitHub

Research Interests

Computer Vision, Reinforcement Learning, Vision-Language Models, Neural Fields, 3D Perception, Egocentric Perception, Embodied AI, Vision-Language-Action Models, Robot Learning

News

May 2026	MolmoSpaces was accepted at RSS 2026! See you in Sydney! 🇦🇺
Apr 2026	New preprint: WHOLE-MoMa: Whole-Body Mobile Manipulation using Offline Reinforcement Learning on Sub-optimal Controllers.
Mar 2026	New preprint: MolmoB0T: Large-Scale Simulation Enables Zero-Shot Manipulation.
Feb 2026	New preprint: MolmoSpaces: Large-Scale Open Ecosystem for Robot Manipulation and Navigation.
Jan 2026	UniFField was accepted at ICRA 2026! See you in Vienna! 🇦🇹
Sep 2025	2HandedAfforder was a Best Paper Award (Finalist) 🏆 at the Human to Robot (H2R) Workshop at CoRL 2025!
Sep 2025	New preprint: UniFField: A Generalizable Unified Neural Feature Field for Visual, Semantic, and Spatial Uncertainties in Any Scene.
Jul 2025	2HandedAfforder & 6DOPE-GS accepted at ICCV 2025! See you in Hawaii! 🌴
Jun 2025	Organizing the EgoAct Workshop on Egocentric Perception & Action for Robot Learning at RSS 2025.
May 2025	Started as a Research Scientist Intern (PhD) at the Allen Institute for AI (Ai2) with the PRIOR team (CV & Embodied AI) in Seattle! 🇺🇸
Mar 2025	New preprint: 2HandedAfforder: Learning Precise Actionable Bimanual Affordances from Human Videos.
Dec 2024	New preprint: 6DOPE-GS: Online 6D Object Pose Estimation using Gaussian Splatting.
May 2024	NeuGraspNet accepted at RSS 2024! See you in Delft! 🇳🇱
May 2024	Organizing the 2nd Workshop on Mobile Manipulation & Embodied Intelligence at ICRA 2024.
Feb 2024	ActPerMoMa accepted at ICRA 2024! See you in Yokohama! 🇯🇵
Jun 2023	Presenting NeuGraspNet as a spotlight at the CVPR Workshop on 3D Vision & Robotics.
May 2023	Organizing the Workshop on Representations, Abstractions, and Priors for Robot Learning (RAP4Robots) at ICRA 2023 in London! 🇬🇧
Oct 2022	Presented & won a Best Mobile Manipulation Paper Award 🏆 for “Robot Learning of Mobile Manipulation with Reachability Behavior Priors” at IROS 2022!
Jun 2022	“Robot Learning of Mobile Manipulation with Reachability Behavior Priors” accepted at IEEE RA-L & IROS 2022! See you in Kyoto! ⛩️

Publications

	WHOLE-MoMa: Whole-Body Mobile Manipulation using Offline Reinforcement Learning on Sub-optimal Controllers Snehal Jauhri, Vignesh Prasad, Georgia Chalvatzaki arXiv preprint, 2026 project page / arXiv
	MolmoB0T: Large-Scale Simulation Enables Zero-Shot Manipulation Snehal Jauhri^, Abhay Deshpande^, Maya Guru^, Rose Hendrix^, Ainaz Eftekhar, Rohun Tripathi, Max Argus, Jordi Salvador, Haoquan Fang, Matthew Wallingford, Wilbert Pumacay, Yejin Kim, Quinn Pfeifer, Ying-Chun Lee, Piper Wolters, Omar Rayyan, Mingtong Zhang, Jiafei Duan, Karen Farley, Winson Han, Eli Vanderbilt, Dieter Fox, Ali Farhadi, Georgia Chalvatzaki, Dhruv Shah, Ranjay Krishna arXiv preprint, 2026. project page / arXiv / code
	MolmoSpaces: Large-Scale Open Ecosystem for Robot Manipulation and Navigation Yejin Kim, Wilbert Pumacay, Omar Rayyan, Max Argus, Winson Han, Eli VanderBilt, Jordi Salvador, Abhay Deshpande, Rose Hendrix, Snehal Jauhri, Shuo Liu, Nur Muhammad Mahi Shafiullah, Maya Guru, Arjun Guru, Ainaz Eftekhar, Karen Farley, Donovan Clay, Jiafei Duan, Piper Wolters, Alvaro Herrasti, Ying-Chun Lee, Georgia Chalvatzaki, Yuchen Cui, Ali Farhadi, Dieter Fox, Ranjay Krishna Robotics: Science and Systems (RSS), 2026 project page / arXiv / code
	UniFField: A Generalizable Unified Neural Feature Field for Visual, Semantic, and Spatial Uncertainties in Any Scene Snehal Jauhri^, Christian Maurer^, Sophie Lueth, Georgia Chalvatzaki IEEE International Conference on Robotics and Automation (ICRA), 2026. Also presented at the ICCV 2025 Open-World 3D Scene Understanding Workshop. project page / arXiv
	2HandedAfforder: Learning Precise Actionable Bimanual Affordances from Human Videos Snehal Jauhri^, Marvin Heidinger^, Vignesh Prasad, Georgia Chalvatzaki IEEE/CVF International Conference on Computer Vision (ICCV), 2025. Also presented at the RSS 2025 EgoAct Workshop and the CoRL 2025 Human to Robot Workshop (Best Paper Award Finalist). project page / arXiv / code
	6DOPE-GS: Online 6D Object Pose Estimation using Gaussian Splatting Yufeng Jin, Vignesh Prasad, Snehal Jauhri, Mathias Franzius, Georgia Chalvatzaki IEEE/CVF International Conference on Computer Vision (ICCV), 2025. Also presented at the RSS 2025 Gaussian Representations Workshop. project page / arXiv
	NeuGraspNet: Learning Any-View 6DoF Robotic Grasping in Cluttered Scenes via Neural Surface Rendering Snehal Jauhri, Ishikaa Lunawat, Georgia Chalvatzaki Robotics: Science and Systems (RSS), 2024. Also presented as a spotlight at the CVPR 2023 Workshop on 3D Vision & Robotics. project page / arXiv / code
	ActPerMoMa: Active-Perceptive Motion Generation for Mobile Manipulation Snehal Jauhri^, Sophie Lueth^, Georgia Chalvatzaki IEEE International Conference on Robotics and Automation (ICRA), 2024. project page / arXiv / code
	Robot Learning of Mobile Manipulation with Reachability Behavior Priors Snehal Jauhri, Jan Peters, Georgia Chalvatzaki IEEE Robotics and Automation Letters (RA-L), vol. 7, no. 3, pp. 8399–8406, 2022. Presented at IROS 2022 (Best Mobile Manipulation Paper Award). project page / arXiv / code
	Safe Reinforcement Learning of Dynamic High-Dimensional Robotic Tasks: Navigation, Manipulation, Interaction Puze Liu, Kuo Zhang, Davide Tateo, Snehal Jauhri, Zhiyuan Hu, Jan Peters, Georgia Chalvatzaki IEEE International Conference on Robotics and Automation (ICRA), 2023. project page / arXiv / code
	Regularized Deep Signed Distance Fields for Reactive Motion Generation Puze Liu, Kuo Zhang, Davide Tateo, Snehal Jauhri, Jan Peters, Georgia Chalvatzaki IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2022. project page / arXiv / code
	Interactive Imitation Learning in State-Space Snehal Jauhri, Carlos Celemin, Jens Kober Conference on Robot Learning (CoRL), vol. 155, pp. 682–692, PMLR, 2020. arXiv / code

Template credits: Jon Barron. Spouse credits: Brinda Mohan.