Research

My work focuses on robot adaptation for iterative improvement, continual learning, and productive human-robot interactions. Refer to my Google Scholar for an up-to-date list of publications.


Will People Enjoy a Robot Trainer? A Case Study with Snoopie the Pacerbot

Maximilian Du*, Jenn Grannen*, Shuran Song, Dorsa Sadigh

International Conference on Robotics and Automation (ICRA) 2026

Simple & Short: The physical embodiment of a robot quadruped makes it good for natural human-robot interactions, including being a personal running trainer.

Website Paper

DynaGuide: Steering Diffusion Polices with Active Dynamic Guidance

Maximilian Du, Shuran Song

Neural Information Processing Systems (NeurIPS) 2025

Simple & Short: We can steer the behavior of a trained robot policy by influencing its action generation process with signals from a model that predicts future outcomes.

Website Paper Code

To Err is Robotic: Rapid Value-Based Trial-and-Error during Deployment

Maximilian Du, Sasha Khazatsky, Tobias Gerstenberg, Chelsea Finn

Simple & Short: We make a failure detector for robots that allow them to detect mistakes and try something else.

Website Paper

Behavior Retrieval: Few-Shot Imitation Learning by Querying Unlabeled Datasets

Maximilian Du, Suraj Nair, Dorsa Sadigh, Chelsea Finn

Robotics: Science and Systems 2023

Simple & Short: We create a data filtering approach that allows a new robot task to take advantage of old robot data.

Website Paper Code

Play it by Ear: Learning Skills amidst Occlusion through Audio-Visual Imitation Learning

Maximilian Du*, Olivia Lee*, Suraj Nair, Chelsea Finn

Robotics: Science and Systems 2022

Simple & Short: We add a microphone to a robot gripper, allowing it to manipulate objects in confined spaces.

Website Paper Code