I am pursuing a Ph.D. in Computer Science at The University of Texas at Austin where I am fortunate to be advised by Prof. Roberto Martín-Martín and am part of the Robot Interactive Intelligence (RobIn) lab. I am passionate about developing general-purpose, deployable robots in the wild. My research focuses on failure-aware and safe robot learning, enabling robots to act reliably in the real world.

I did a Research Internship at the Robotics and AI Institute (RAI) in Summer 2025. I previously finished my MS in Computer Science from Columbia University, advised by Prof. Shuran Song, and my undergraduate from the Indian Institute of Information Technology - Allahabad (IIIT-A), advised by Prof. G. C. Nandi.

I also love to play tennis, basketball, music and to snowboard.

Arpit Bahety*, Skand Peri*, Sriniket Ambatipudi, Chen Xu, Haruki Nishimura, Masha Itkina, Roberto Martín-Martín
Preprint 2026
Website  •   Paper

TL;DR: Enables robots to learn to recover from failures using very few corrections by levaraging semantics.

Arnav Balaji*, Arpit Bahety*, Sriniket Ambatipudi, Daniel Lam, Junhong Xu, Roberto Martín-Martín,
Robotics: Science and Systems (RSS 2026)
Website  •   Paper   •   Code

TL;DR: A damage-aware, simulator-agnostic framework and benchmark for evaluating and learning safer policies.

Chengshu Li*, Mengdi Xu*, Arpit Bahety*, Hang Yi*, Yunfan Jiang, Huang Huang, Josiah Wong, Sujay Garlanka, Cem Gokmen, Ruohan Zhang, Weiyu Liu, Jiajun Wu, Roberto Martín-Martín, Fei-Fei Li
International Conference on Learning Representations (ICLR 2026)
Website  •   Paper   •   Video   •   Code   •   Poster

TL;DR: An automated data generation method for multi-step bimanual mobile manipulation that satisfies hard constraints (e.g., reachability) while balancing soft constraints (e.g., visibility).

Arpit Bahety, Arnav Balaji, Ben Abbatematteo, Roberto Martín-Martín
Robotics: Science and Systems (RSS 2025)
Website  •   Paper   •   Video   •   Poster   •   RSS Presentation

TL;DR: A frameowrkd that enables robots to safely and autonomously learn multi-step mobile manipulation tasks from a single human video by segmenting, translating, and adapting the demonstrated actions to their own morphology.

Arpit Bahety, Priyanka Mandikal, Ben Abbatematteo, Roberto Martín-Martín
Robotics: Science and Systems (RSS 2024)
Outstanding Student Paper Finalist
Website  •   Paper   •   Video   •   Code   •   Poster   •   RSS Presentation

TL;DR: A method to teach robots bimanual manipulation using human video demonstrations, by leveraging a novel, psychology- and biomechanics-inspired action space to learn effectively from failures in the real world.

Gu-Cheol Jeong, Arpit Bahety, Gabriel Pedraza, Ashish D. Deshpande, Roberto Martín-Martín
International Conference on Intelligent Robots and Systems (IROS 2024)
Best Paper Award on Robot Mechanisms and Designs
Website  •   Paper   •   Assembly Instructions   •   Usage Code   •   Video

TL;DR: A novel hybrid rigid–flexible robotic gripper that enhances versatility (e.g. compliant grasping and precise pinching) and collision-tolerance, enabling safer and robust robot learning in everyday environments.

Zeyi Liu*, Arpit Bahety*, Shuran Song
Conference on Robot Learning (CoRL 2023)
Website  •   Paper   •   Code   •   Video

TL;DR: A framework that leverages LLM for robot failure explanation and correction, based on a hierarchical summary of robot past experiences generated from multisensory data.

Arpit Bahety*, Shreeya Jain*, Huy Ha, Nathalie Hager, Benjamin Burchfiel, Eric Cousineau, Siyuan Feng, Shuran Song
International Conference on Intelligent Robots and Systems (IROS 2023)
Website  •   Paper   •   Code (coming soon)   •   Video

TL;DR: A robotic system that learns to pack rigid and deformable objects into deformable bags using two learned policies for object rearrangement and bi-manual bag lifting.

Arpit Bahety, Rohit Saluja, Ravi Kiran Sarvadevabhatla, Anbumani Subramanian, C. V. Jawahar
Indian Conference on Computer Vision, Graphics and Image Processing (ICVGIP), 2021
Paper   •   Code   •   Video   •   The Indian Express   •   Telangana Today   •   Edexlive

Arpit Bahety, Ana-Maria Olteteanu
7th International Workshop on Artificial Intelligence and Cognition, 2019
Paper

Ana-Maria Olteteanu, Mikkel Schöttner, Arpit Bahety
Frontiers in Psychology, 2019
Paper

* indicates equal contribution

• Conference Reviewer: ICRA, IROS
• Teaching Assistant:
  • COMS 4773: Robot Manipulation and Learning (Spring 2024, UT Austin)
  • COMS 4773: Computational Aspects of Robotics (Spring 2022 and Spring 2023, Columbia University)
  • COMS 4995: Neural Networks and Deep Learning (Fall 2021, Columbia University)