Implicit-Behavior Coordination from Unlabeled Sub-Task Demonstrations for Rearrangement Tasks

2026-07-10

AI Summary

Long-horizon robotic rearrangement tasks typically rely on explicit skill abstractions—such as predefined skills or labels—which struggle to scale with increasing task length and behavioral complexity. This work proposes a purely data-driven, implicit behavior coordination mechanism that learns skill-like behaviors directly from unlabeled, mixed-subtask demonstrations, without requiring prior task planning knowledge or explicit skill annotations. By integrating behavior cloning with value-function-guided candidate action selection, the method enables adaptive coordination of multimodal behaviors. Evaluated on Habitat rearrangement tasks, the approach significantly outperforms task-specific imitation baselines, achieves performance approaching that of an oracle planner using behavior cloning skills, and demonstrates superior robustness when scaling to longer tasks and larger behavior repertoires.

Abstract

Long-horizon robotic rearrangement tasks are often treated as skill sequencing problems, requiring predefined skills, skill labels, or boundaries, and task-specific switching logic. Although effective, such explicit skill abstractions can become difficult to scale as the number of behaviors and the task horizon increase. We instead formulate rearrangement as implicit-behavior coordination from unlabeled sub-task demonstrations, where skill-like behaviors are learned directly from mixed behavior data and coordinated through value-guided action selection. Experiments in Habitat rearrangement tasks support this formulation in three ways. First, our method outperforms task-specific imitation baselines on more complex rearrangement tasks and approaches an oracle-planner baseline with behavior-cloned skills, while using no oracle task plan or skill-labeled full-task demonstrations. Second, ablations show that reliable critic-guided candidate selection is essential for coordinating multi-modal behaviors. Third, scaling experiments show that the method handles larger behavior repertoires and maintains stronger performance than task-specific imitation baselines as chained targets extend the horizon. These results suggest that explicit skill abstraction is not a prerequisite for long-horizon rearrangement, and that implicit-behavior coordination offers a promising data-driven alternative to explicit skill-based pipelines.

Problem

• implicit-behavior coordination
• unlabeled sub-task demonstrations
• rearrangement tasks
• long-horizon robotic tasks
• skill abstraction

Innovation

• implicit-behavior coordination
• unlabeled sub-task demonstrations
• value-guided action selection
• long-horizon rearrangement
• skill-free learning