通过习惯性动作规则减少大模型智能体的重复调用，提升长周期任务执行效率。
HaReCAP: Habitual-action Grounding for Recursive Large Language Model Agents
- 将高频成功动作提炼为离线规则，替代部分LLM调用进行底层动作生成。
- 在多个环境上降低14.67%至20.08%的令牌消耗，效果稳定。
- 无需修改原有框架，适合希望优化推理开销的研究者与开发者。
长周期具身任务要求大语言模型智能体迭代分解高层目标、根据环境反馈修正计划，并将最底层子目标映射为有效可执行动作。递归上下文管理方法如ReCAP通过多级任务分解和父节点优化提升规划稳定性，但仍需在叶节点反复调用LLM以将原子子任务转化为精确合法动作。我们称此最终映射步骤为‘最后一公里接地冗余’，在长周期执行中累积导致大量LLM调用与令牌开销。为此，我们提出HaReCAP（Habitual-action Grounded ReCAP），一种对ReCAP的低侵入式叶节点接地扩展。HaReCAP从成功轨迹中提取频繁出现的叶级决策，离线编译为可审计且可放弃的一步反射规则。运行时，若规则能唯一确定当前合法动作集中的有效动作，则跳过叶节点LLM调用；否则回退至原ReCAP流程。该设计避免了将完整递归上下文反复传入LLM处理常规动作接地，同时保留原有递归控制流。我们在Robotouille和ALFWorld上以Qwen3.5-27B为主模型进行评估。在两种方法均能解决的任务上，HaReCAP分别在Robotouille同步、异步和ALFWorld场景下减少14.67%、17.93%和20.08%的令牌消耗。结果表明，HaReCAP可作为ReCAP类递归上下文管理框架的低侵入扩展，在不同环境与模型上有效缓解最后一公里接地冗余，适用于常见成功轨迹。
原文摘要 · Abstract (English)
Long-horizon embodied tasks require LLM agents to iteratively decompose high-level goals, revise plans in response to environmental feedback, and ground leaf-level subgoals into valid executable actions. Recursive context-management methods such as ReCAP improve planning stability through multi-level task decomposition and parent-node refinement, but still repeatedly invoke the LLM at leaf nodes to ground atomic subtasks into exact valid actions. We refer to this final grounding step as last-mile grounding redundancy, which accumulates into substantial LLM-call and token overhead during long-horizon execution. To mitigate this issue, we propose HaReCAP (Habitual-action Grounded ReCAP), a low-intrusion leaf grounding extension for ReCAP. HaReCAP extracts frequent leaf decisions from successful trajectories and compiles them offline into auditable and abstainable one-step leaf-reflex rules. At runtime, it skips the leaf LLM call only when a rule can uniquely determine a legal action in the current valid-action set; otherwise, it falls back to the original ReCAP. This design avoids repeatedly carrying the full recursive context into the LLM for routine leaf action grounding, while preserving the original recursive control flow. We evaluate HaReCAP on Robotouille and ALFWorld with Qwen3.5-27B as the main model. On tasks solved by both ReCAP and HaReCAP, HaReCAP reduces token consumption by 14.67%, 17.93%, and 20.08% on Robotouille synchronous, Robotouille asynchronous, and ALFWorld, respectively. The results show that HaReCAP can serve as a low-intrusion extension to ReCAP-style recursive context-management frameworks, reducing last-mile grounding redundancy across environments and models on commonly successful trajectories.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。