从智能体轨迹中提炼通用推理模块，显著提升任务表现。
Inducing Reasoning Primitives from Agent Traces
- 通过聚类分析成功推理路径，提取高频动作形成可复用的伪工具库。
- 在多个任务上性能超越原始智能体，最高提升44个百分点。
- 适合需要高效推理的复杂规划任务，尤其适合资源受限场景。
ReAct风格的大模型智能体常重复发现相同的推理模式，但这些模式仅存在于临时草稿中。本文提出推理原语归纳方法，仅需一次遍历即可挖掘成功的ReAct轨迹，聚类重复的推理步骤，并将最频繁的动作转化为一个紧凑的带类型伪工具库。每个伪工具由自然语言文档字符串定义，由LLM在调用时解析；标准ReAct循环在测试时组合这些原语。核心结果表明，归纳出的工具库性能优于生成其轨迹的原始智能体：在RuleArena NBA任务上提升44个百分点（30 → 74），在MuSR团队分配任务上提升30个百分点（38 → 68），在NatPlan会议规划任务上提升22个百分点（7 → 29）。在五个涵盖叙事推断、规则应用与约束满足规划的子任务中，单一固定配置在所有任务上均优于零样本链式思考，达到或超过专家编写的分解方案，且以更低平均推理成本超越AWM。
原文摘要 · Abstract (English)
ReAct-style LLM agents often rediscover the same reasoning routines across problems, yet leave those routines trapped in transient scratchpads. We introduce Reasoning Primitive Induction, a single-pass method that mines successful ReAct traces, clusters recurrent reasoning moves, and converts the most frequent moves into a compact library of typed pseudo-tools. Each pseudo-tool is specified by a natural-language docstring interpreted by an LLM at invocation time, and a standard ReAct loop composes these primitives at test time. The central result is that induced libraries outperform the very agent that generated their traces: by +44pp on RuleArena NBA (30 -> 74), +30pp on MuSR team allocation (38 -> 68), and +22pp on NatPlan meeting planning (7 -> 29). Across five comparable subtasks spanning narrative deduction, rule application, and constraint-satisfaction planning, a single fixed configuration improves over zero-shot Chain-of-Thought on every subtask, matches or surpasses expert-authored decompositions, and outperforms AWM at lower average inference cost.
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。