arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2601.07577cs.AI
收藏

超越纠缠规划:面向长时间 horizon 的任务解耦规划

Beyond Entangled Planning: Task-Decoupled Planning for Long-Horizon Agents

State Key Laboratory of AI Safety, Institute of Computing Technology, CAS(人工智能安全国家重点实验室、计算技术研究所、中国科学院);University of Chinese Academy of Sciences(中国科学院大学)

Yunfan Li, Bingbing Xu, Xueyun Tian, Xiucheng Xu, Huawei Shen

AI总结:

本文提出任务解耦规划(TDP)方法,通过将任务分解为子目标图,减少长horizon智能体的规划误差传播,提升鲁棒性和效率。

摘要:

近年来,大语言模型(LLMs)的进步使智能体能够自主执行复杂且长时间 horizon 的任务,但规划仍然是可靠任务执行的主要瓶颈。现有方法通常分为两种范式:分步规划,其反应性强但往往目光短浅;以及一次性规划,其会提前生成完整计划但容易因执行错误而崩溃。关键的是,这两种范式都存在纠缠上下文的问题,即智能体必须在跨越多个子任务的单一历史中进行推理。这种纠缠增加了认知负担,使局部错误在其他独立决策中传播,使恢复变得计算昂贵。为了解决这一问题,我们提出了任务解耦规划(TDP),一种无需训练的框架,它用任务解耦取代纠缠推理。TDP通过监督将任务分解为有向无环图(DAG)的子目标。使用具有受限上下文的规划器和执行器,TDP将推理和重新规划限制在活跃的子任务中。这种隔离防止了错误传播,并能局部纠正偏差而不影响工作流程。在TravelPlanner、ScienceWorld和HotpotQA上的结果表明,TDP在性能上优于强大的基线模型,同时将token消耗减少了高达82%,证明了子任务解耦在提高长horizon智能体的鲁棒性和效率方面的有效性。

英文摘要:

Recent advances in large language models (LLMs) have enabled agents to autonomously execute complex, long-horizon tasks, yet planning remains a primary bottleneck for reliable task execution. Existing methods typically fall into two paradigms: step-wise planning, which is reactive but often short-sighted; and one-shot planning, which generates a complete plan upfront yet is brittle to execution errors. Crucially, both paradigms suffer from entangled contexts, where the agent must reason over a monolithic history spanning multiple sub-tasks. This entanglement increases cognitive load and lets local errors propagate across otherwise independent decisions, making recovery computationally expensive. To address this, we propose Task-Decoupled Planning (TDP), a training-free framework that replaces entangled reasoning with task decoupling. TDP decomposes tasks into a directed acyclic graph (DAG) of sub-goals via a Supervisor. Using a Planner and Executor with scoped contexts, TDP confines reasoning and replanning to the active sub-task. This isolation prevents error propagation and corrects deviations locally without disrupting the workflow. Results on TravelPlanner, ScienceWorld, and HotpotQA show that TDP outperforms strong baselines while reducing token consumption by up to 82%, demonstrating that sub-task decoupling improves both robustness and efficiency for long-horizon agents.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑