ALAS: Adaptive Long-Horizon Action Synthesis via Async-pathway Stream Disentanglement

arXiv:2604.20721v1 Announce Type: new 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 ALAS, a cross-domain learning framework for LH tasks via biologically inspired dual-stream disentanglement. Inspired by the brain's "where-what" dual pathway mechanism, ALAS 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, ALAS can achieve an average subtasks success rate improvement of 23% and average execution efficiency improvement of 29%.

Related

• Hierarchical Reinforcement Learning with Augmented Step-Level Transitions for LLM Agents
• A1: A Fully Transparent Open-Source, Adaptive and Efficient Truncated Vision-Language-Action Model
• Test-Time Perturbation Learning with Delayed Feedback for Vision-Language-Action Models
• SYMBOLIZER: Symbolic Model-free Task Planning with VLMs
• A Hierarchical Spatiotemporal Action Tokenizer for In-Context Imitation Learning in Robotics

Source: arXiv cs.RO | 2026-04-23