大型语言模型越来越多地被部署为能够进行长期规划并通过外部工具执行操作的代理。大多数代理通过在累积的历史,上不受约束的生成来选择操作，从而隐含了要做什么,以什么顺序,以及在什么条件下做什么的程序知识。随着轨迹延长, 代理可能会失去目标, 调用无序的工具, 并重复无效的操作。我们引入过程图:，就像知识图将事实知识组织成(实体,关系,实体)三元组，解决“是什么”问题,，过程图将过程知识组织成(过程,关系,过程)三元组，解决做什么问题。在每个决策步骤,，框架定位代理的 活动节点,，并且指导模型将周围的子图转换为步骤级情景指导，该指导对求解器的 的下一步行动进行偏置而不指示它。该图是自我演化的: LLM 细化器将失败的轨迹与成功的轨迹进行对比，并编辑图的 拓扑和属性, 提交编辑以保留或改进保留的验证性能，同时保留被拒绝的轨迹以阻止重复。从最小的骨架,开始，循环构建匹配或超过手工设计的图表。它还可以事先修复有缺陷的专家。在多个数据集, 任务类型, 和LLM, 中，程序图比基于内存的基线, 提供了一致的增益，并且自我进化进一步提高了性能，无需手动工程。

Large language models are increasingly deployed as agents that plan over long horizons and act through external tools. Most agents select actions through unconstrained generation over an accumulating history, leaving implicit the procedural knowledge of what to do, in what order, and under which conditions. As trajectories lengthen, agents can lose track of their objectives, invoke tools out of order, and repeat unproductive actions. We introduce the Procedural Graph: just as a knowledge graph organizes factual knowledge into (entity, relation, entity) triplets for what-is questions, a Procedural Graph organizes procedural knowledge into (procedure, relation, procedure) triplets for what-to-do questions. At each decision step, the framework localizes the agent的 active node, and a guidance model translates the surrounding subgraph into step-level situational guidance that biases the solver的 next action without dictating it. The graph is self-evolving: an LLM refiner contrasts failed trajectories with successful ones and edits the graph的 topology and attributes, committing edits that preserve or improve held-out validation performance while retaining rejected ones to discourage repetition. Starting from a minimal skeleton, the loop builds graphs that match or surpass hand-designed ones. It can also repair a flawed expert prior. Across multiple datasets, task types, and LLMs, the Procedural Graph delivers consistent gains over memory-based baselines, and self-evolution further improves performance without manual engineering.

科目: 人工智能 (cs.AI); 计算和语言 (cs.CL); 多代理系统 (cs.MA)

Subjects: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Multiagent Systems (cs.MA)