SID: Sliding into Distribution for Robust Few-Demonstration Manipulation

2026-05-13 · Yicheng Ma, Wei Yu, Zhian Su, Xidan Zhang, Huixu Dong arxiv

Generalizing robotic manipulation across object poses, viewpoints, and dynamic disturbances is difficult, especially with only a few demonstrations. End-to-end visuomotor policies are expressive but data-hungry, while planning and optimization satisfy explicit constraints but do not directly capture the interaction strategies demonstrated by humans. We propose Sliding into Distribution (SID), a structured framework that learns an object-centric motion field from canonicalized demonstrations to iteratively slide the system toward the demonstrated manifold and into the reliable operating region of a lightweight egocentric execution policy, mitigating out-of-distribution (OOD) execution. The motion field provides large corrective motions when far from the demonstration manifold and naturally vanishes near convergence, enabling robust reaching under substantial pose and viewpoint shifts. Within the reached regime, an egocentric policy trained with conditioned flow matching performs task-specific manipulation, supported by kinematically consistent point-cloud reprojection augmentation that preserves action-observation consistency. Across six real-world tasks, SID achieves approximately 90% success under OOD initializations with only two demonstrations, with under a 10% drop under distractors and external disturbances. Overall, SID provides a new paradigm for few-shot manipulation: explicitly managing distribution shift via online distribution recovery.

📄 PDF Abstract BibTeX arXiv:2605.13428

Code (0)

등록된 구현이 없습니다.

Learning to Compose Hierarchical Object-Centric Controllers for Robotic Manipulation

2020-11-09 · Mohit Sharma, Jacky Liang, Jialiang Zhao, Alex LaGrassa 외

Manipulation tasks can often be decomposed into multiple subtasks performed in parallel, e.g., sliding an object to a goal pose while maintaining contact with a table. Individual subtasks can be achieved by task-axis con…

Object reinforcement-learning Reinforcement Learning (RL)Zero-shot Generalization

Evaluating the Effectiveness of Corrective Demonstrations and a Low-Cost Sensor for Dexterous Manipulation

2022-04-15 · Abhineet Jain, Jack Kolb, J. M. Abbess IV, Harish Ravichandar

Imitation learning is a promising approach to help robots acquire dexterous manipulation capabilities without the need for a carefully-designed reward or a significant computational effort. However, existing imitation le…

Imitation Learning

One Demonstration Is Enough for Real-World Robotic Reinforcement Learning

2026-07-02 · Yuwan Liu, Hongze Yu, Song Liu, Yuhan Wang 외 arxiv

Learning effective robot control policies on physical hardware is challenging due to costly data collection and the difficulty of reward specification. Prior work has incorporated demonstrations into reinforcement learni…

Reinforcement Learning

From a Single Demonstration to a General Policy for Contact-Rich Manipulation

2026-05-17 · Xing Li, Oliver Brock arxiv

We present a Learning from Demonstration (LfD) framework that achieves one-shot generalization in multi-stage, contact-rich manipulation tasks. Central to our approach is the utilization of environmental constraints as t…

GHOST: Hierarchical Sub-Goal Policies for Generalizing Robot Manipulation

2026-06-08 · Sriram Krishna, Ben Eisner, Haotian Zhan, Ying Yuan 외 arxiv

We present GHOST, a framework for learning visuomotor manipulation policies that generalize beyond the training distribution. GHOST factorizes control into (i) a high-level policy that predicts the next sub-goal as a dis…

Robot Manipulation