Nitish Dashora

Nitish Dashora

he/him/his

AI Researcher

Publications

You can also find my articles on my Google Scholar profile.

Conference Papers

Correct Me If I’m Wrong: Language Guided Exploration for VLAs

Published in Reinforcement Learning for Vision-Language-Action Models Workshop, 2026

We let humans steer VLA exploration during post-training by simply revising the language instruction.

Recommended citation: Sunshine Jiang, John Marangola, David Zhang, Nitish Dashora, Pulkit Agrawal, Zhang-Wei Hong. (2026). "Correct Me If I’m Wrong: Language Guided Exploration for VLAs." Reinforcement Learning for Vision-Language-Action Models Workshop.
Download Paper

Learning More from Less: Reinforcement Learning from Hindsight

Published in Reinforcement Learning for Vision-Language-Action Models Workshop, 2026

We relabel failed robot rollouts with the tasks they actually achieved to make RL post-training of VLAs far more sample-efficient.

Recommended citation: Iris Xu, Sunshine Jiang, John Marangola, Nitish Dashora, Richard Li, Thomas Liu, Zexue He, Yuheng Zhi, Alex Pentland, Pulkit Agrawal, Zhang-Wei Hong. (2026). "Learning More from Less: Reinforcement Learning from Hindsight." Reinforcement Learning for Vision-Language-Action Models Workshop.
Download Paper

Prompt-Driven Exploration

Published in ArXiv, 2026

We explore in prompt space, letting a VLM rewrite instructions from failed rollouts to escape weak policies.

Recommended citation: Sunshine Jiang, John Marangola, David Zhang, Raghuram Kowdeed, Ruiyang Luo, Nitish Dashora, Richard Li, Pulkit Agrawal, Zhang-Wei Hong. (2026). "Prompt-Driven Exploration." arXiv preprint arXiv:2607.08837.
Download Paper

ViNT: A Foundation Model for Visual Navigation

Published in Conference on Robot Learning (Oral), 2023

We present a multi-robot visual navigation foundation model for effective fine-tuning and novel task adaptation.
Also appeared as: Oral Presentation at BayLearn, Live Demo at CoRL, Workshop Paper at NeurIPS, YouTube Video, and Project Site

Recommended citation: Dhruv Shah*, Ajay Sridhar*, Nitish Dashora*, Kyle Stachowicz, Kevin Black, Noriaki Hirose, Sergey Levine. (2023). "ViNT: A Large-Scale, Multi-Task Visual Navigation Backbone with Cross-Robot Generalization." In the proceedings of 2023 Conference on Robot Learning (CoRL).
Download Paper

Imitative Models for Passenger-Scale Autonomous Off-Road Driving

Published in IPPC, LRSA @ International Conference on Intelligent Robots and Systems, 2023

DAgger-like imitative learning with model predictive path integral control works as an end-to-end navigation system.
Also appeared as: YouTube Video

Recommended citation: Nitish Dashora, Sunggoo Jung, Dhruv Shah, Valentin Ibars, Osher Lerner, Chanyoung Jung, Rohan Thakker, Nicholas Rhinehart, Ali-akbar Agha-mohammadi. (2022). "Imitative Models for Passenger-Scale Autonomous Off-Road Driving." Presented at 2023 International Conference on Intelligent Robots and Systems (IROS).
Download Paper

Hybrid Imitative Planning with Geometric and Predictive Costs in Off-road Environments

Published in International Conference on Robotics and Automation, 2022

Combining geometric costmaps and learned models improves long-range navigation in offroad environments.
Also appeared as: Workshop Paper at NeurIPS, YouTube Video, and Project Site

Recommended citation: Nitish Dashora*, Daniel Shin*, Dhruv Shah, Henry Leopold, David Fan, Ali Agha-Mohammadi, Nicholas Rhinehart, Sergey Levine. (2022). "Hybrid Imitative Planning with Geometric and Predictive Costs in Off-road Environments." In the proceedings of 2022 International Conference on Robotics and Automation (ICRA).
Download Paper