DETACH: Cross-domain Learning for Long-Horizon Tasks via Mixture of Disentangled Experts

TL;DR

DETACH is a biologically inspired framework that enables cross-domain learning for long-horizon tasks by disentangling environment understanding and skill execution, improving generalization and efficiency.

Contribution

The paper introduces DETACH, a novel dual-stream disentanglement approach for cross-domain long-horizon tasks, inspired by brain mechanisms, enabling better transfer and generalization.

Findings

Achieves 23% higher subtask success rate

Improves execution efficiency by 29%

Effective across diverse human-scene interaction tasks

Abstract

Long-Horizon (LH) tasks in Human-Scene Interaction (HSI) are complex multi-step tasks that require continuous planning, sequential decision-making, and extended execution across domains to achieve the final goal. However, existing methods heavily rely on skill chaining by concatenating pre-trained subtasks, with environment observations and self-state tightly coupled, lacking the ability to generalize to new combinations of environments and skills, failing to complete various LH tasks across domains. To solve this problem, this paper presents DETACH, a cross-domain learning framework for LH tasks via biologically inspired dual-stream disentanglement. Inspired by the brain's "where-what" dual pathway mechanism, DETACH comprises two core modules: i) an environment learning module for spatial understanding, which captures object functions, spatial relationships, and scene semantics, achieving cross-domain transfer through complete environment-self disentanglement; ii) a skill learning module for task execution, which processes self-state information including joint degrees of freedom and motor patterns, enabling cross-skill transfer through independent motor pattern encoding. We conducted extensive experiments on various LH tasks in HSI scenes. Compared with existing methods, DETACH can achieve an average subtasks success rate improvement of 23% and average execution efficiency improvement of 29%.