Jin-Yan Liu

LM-X: Explainable Action Modeling with Progress, Event, and Uncertainty Prediction for Generalist Robot Manipulation

KungfuAthleteBot: learning high-dynamic humanoid motion from video with unified robust recovery

Video is an abundant, inexpensive source of human motion data that is rich in extreme athletic behaviors. Making it usable for humanoid robots, however, is not a matter of simply retargeting a reconstructed trajectory: video-derived motion is physically inconsistent, devoid of actuation information, and says nothing ab...

ForgetMimic: Motion Unlearning for Reinforcement Learning Humanoid Control

Experimental results demonstrate that ForgetMimic effectively eliminates memory of designated motions while maintaining the normal operation of all other motions, and identifies and resolves two key training mechanisms in robot control that lead to unlearning failure.

ViTaR: Visuo-Tactile Residual Adaptation for Foundation VLA Manipulation

ViTaR is introduced, which reframes tactile feedback from an action-generating perceptual input to an execution modulator that selects and scales bounded residual corrections atop a frozen VLA, preserving pretrained capabilities by construction.

LM-X: Explainable Vision--Language--Action Modeling via Progress, Event, and Uncertainty Prediction

LM-X is introduced, which organizes prediction across task, event, and motor scales without claiming anatomical correspondence and shows that explicit multi-timescale predictive state can strengthen control while exposing interpretable internal estimates.

MemCatalyst: Amplifying Data Auditing on Vision-Language Models via Data Poisoning

This work proposes MemCatalyst, a set of data poisoning tools, aiming to amplify the data auditing performance on VLMs, and forces VLMs to over-learn specific inconsistencies between image features and textual semantics during training, thereby increasing their susceptibility to membership information auditing.