Hu-Chuan Lu

This work proposes CARE (Corrective Atomic Robotic Execution), a framework that improves recovery by learning from failures encountered during execution, and introduces the Failure State Recovery Benchmark (FSR-Bench), which evaluates recovery from intermediate failure states under local deviations and structural anomalies.

The results show that a diffusion policy can benefit substantially from knowing where an interaction is heading, without being told exactly where to move.

VA-Bench is introduced to evaluate the complete observe-reason-act-revise loop and tests whether general-purpose MLLMs can turn visual demonstrations and actively acquired evidence into successful embodied action.

Attention-DP3 is proposed, a spatially object-aware 3D diffusion policy that injects object-level geometric cues via attention while keeping the DP3 diffusion backbone unchanged and shows consistent improvements over DP3, achieving state-of-the-art performance across benchmarks.

A training-free framework that combines a query-independent layout-derived interaction prior with instruction relevance and feature novelty to rank visual evidence according to both potential future utility and diversity, allowing retained visual evidence to shrink monotonically across budgets while remaining reusable throughout the trajectory.

A teacher-student framework reveals that leveraging privileged target appearances can dramatically enhance tracking accuracy, closing a critical performance gap.

MA-VLA achieves robust multi-arm compositional generalization, outperforming existing models by effectively enabling role-agnostic behavior in unseen collaboration scenarios.

GROVE achieves state-of-the-art performance in memory retrieval for wearable assistants by seamlessly integrating reactive and proactive memory access from continuous video streams.