arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2607.01767cs.AI
收藏

修复放大器,而非症状:面向智能体轨迹的稳定世界模型修正

Repair the Core, Not the Symptom: Planning-Graph Repair for Long-Horizon Agents

Emory University(埃默里大学);The University of Tokyo(东京大学);LocationMind

Xinyuan Song, Zekun Cai

AI总结:

针对长流程规划中的图级错误,提出WM-SAR方法,通过子图放大反向定位因果子图进行修复,在有限token预算下优于传统扫描修复方法。

摘要:

随着智能体规划从短工具链转向包含数千或数万步的持久工作流,失败将发生在大型规划图内部而非孤立预测中。每次错误后重新规划整个图既不计算可行也不可取:全图重放消耗大量上下文预算,使LLM暴露于许多无关症状,并可能降低长上下文检索能力。本文研究了此类系统中缺失的组件:一个能够就地修复失败规划图的世界模型修正器。我们比较了两类修正器。第一类是常见的工程方法:扫描节点和边,选择一个可疑的局部区域,并让LLM修复它。我们实现了强大的工程LLM修正器,发现它们能够有所帮助,尤其是在提供非常大的上下文时。第二类是我们的方法WM-SAR(世界模型子图放大修复):它不是扫描可见症状,而是从子图放大反向工作,识别持续放大错误的节点和边,并仅将该因果子图发送给LLM。在图模拟和LLM修复实验中,WM-SAR在现实的token预算下显著优于工程修正器,通过紧凑区域实现接近全图的稳定化,并为LLM提供了更清晰的修复目标。

英文摘要:

Long-horizon language agents make interdependent decisions across multiple rounds, forming growing planning graphs in which later decisions depend on earlier states, observations, and tool outputs. Errors in these graphs can therefore propagate through subsequent decisions and amplify over long rollouts. Correcting the most visible failure may therefore fix a symptom while leaving the core that drives downstream errors unchanged. Replaying the entire planning graph provides the opposite extreme, but becomes increasingly expensive as the graph grows and forces an LLM to reason over substantial irrelevant context. We study planning-graph repair: given a failed rollout, the goal is to identify and correct the compact core responsible for continued error propagation while preserving the remainder of the graph. We first evaluate a range of practical engineering repair strategies, including single-error and TopK repair, temporal windows, neighborhood expansion, cascade-aware repair, and full-graph LLM correction. We then propose ReCore (Repair the Core), which models the coupled propagation of node and edge errors to identify the core of a failed planning graph. Rather than repairing the largest observed errors or expanding a fixed local neighborhood, ReCore selects a connected core whose correction maximally reduces residual error amplification, and exposes only this core to the LLM for diagnosis and repair. We also theoretically establish a bound on long-horizon planning regret. Extensive experiments on synthetic Agent Calling-Tree graphs and the Who\&When attribution benchmark show that ReCore improves long-horizon stabilization and root-cause localization by targeting high-amplification repair regions, matching full-context recovery with up to $7.4\times$ fewer tokens for every successful repair.

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。

↑