7 papers
CARE: Experience-Guided Atomic Corrective Execution for Vision-Language-Action Policies
Junlan Xiao, Junwei Jiang, Zaibin Zhang +4
Vision-Language-Action (VLA) policies achieve strong performance in robotic manipulation but remain brittle once execution deviates from nominal trajectories. We propose CARE (Corr…
Learning Foresight without Explicit Trajectories for 3D Diffusion Policies
Zhongbo Zhang, Zaibin Zhang, Yifan Wang +3
3D diffusion policies are strong at generating geometrically grounded actions from current observations, but successful manipulation requires not only knowing what motion is feasib…
VABench: Measuring Embodied Spatial Intelligence through Visual Demonstrations, Active Perception, and Metric Control
Zhongbo Zhang, Jiayi Jin, Yifan Wang +4
Spatial intelligence requires more than describing object locations. Under incomplete observation, models must identify and acquire missing evidence, interpret it in a common spati…
Attention-DP3: Spatially Object-aware 3D Diffusion Policy via Geometry-aligned Attentional Conditioning
Changbo Yan, Zhongbo Zhang, Zaibin Zhang +3
3D point-cloud observations are inherently ambiguous in complex, cluttered manipulation scenes, where target objects may be partially occluded or tightly intermingled with visually…
MA-VLA: Multi-Arm Vision-Language-Action Model for Collaboration and Compositional Generalization
Zaibin Zhang, Junlan Xiao, Zhongbo Zhang +11
Multi-arm collaboration is becoming a core capability in embodied manipulation. Recent vision-language-action (VLA) models integrate perception, language, and control, but most rep…
Drift and Dependence: Layer-wise Information-Theoretic Bounds for Replay-Based Continual Learning
Tieliang Gong, Zhongbo Zhang, Wen Wen +1
Continual learning must absorb new tasks without erasing old ones, and replay---mixing a small buffer of past examples into current training---is among the most effective remedies…
Think3D: Thinking with Space for Spatial Reasoning
Zaibin Zhang, Yuhan Wu, Lianjie Jia +10
While Vision-Language Models (VLMs) excel at 2D visual understanding, they remain constrained by 2D-centric paradigm that severely limits genuine 3D spatial reasoning. To bridge th…