Principles that Guide, Actions that Inform: Agent Evolution via Knowledge Abstraction

查看原文 / View Original 编辑 / Edit

Computer Science > Artificial Intelligence arXiv:2610.06964 (cs) [Submitted on 3 Oct 2026] Title: Principles that Guide, Actions that Inform: Agent Evolution via Knowledge Abstraction Authors: Bowen Ye, Yongchao Xu, Junkai Ma, Xiang Yin, Wenzhao Li.

计算机科学 > 人工智能 arXiv:2610.06964 (cs) [提交于 2026 年 10 月 3 日] 标题：指导原则与信息行动：通过知识抽象实现智能体进化作者：Bowen Ye, Yongchao Xu, Junkai Ma, Xiang Yin, Wenzhao Li。

Abstract: Large language model (LLM) agents have demonstrated strong capabilities in interactive environments, yet their ability to continually evolve from experience remains limited. Although fine-tuning enables adaptation, its dependence on parameter access and high computational costs restrict its flexibility, especially for large-scale and closed-source LLMs.

摘要：大型语言模型（LLM）智能体在交互式环境中展现出了强大的能力，但它们从经验中持续进化的能力仍然有限。尽管微调可以实现适应性，但其对参数访问的依赖和高昂的计算成本限制了其灵活性，特别是对于大规模和闭源的 LLM 而言。

External memory offers an alternative by allowing agents to accumulate experience without modifying model parameters. However, existing methods mainly focus on experience representation and organization, while the acquired knowledge remains tightly coupled with specific tasks and contexts, limiting generalization.

外部记忆提供了一种替代方案，允许智能体在不修改模型参数的情况下积累经验。然而，现有方法主要关注经验的表示和组织，而所获取的知识仍然与特定任务和上下文紧密耦合，从而限制了泛化能力。

A key challenge is how to transform concrete interactions into abstract and reusable knowledge that guides future decisions beyond individual experiences. To address this challenge, we propose SAGA (Self-evolving Agents through Experience-Grounded Abstraction), a framework for experience-grounded knowledge abstraction and utilization in LLM agents.

一个关键挑战是如何将具体的交互转化为抽象且可重用的知识，以指导超越个体经验的未来决策。为了应对这一挑战，我们提出了 SAGA（通过经验基础抽象实现自我进化的智能体），这是一个用于 LLM 智能体中基于经验的知识抽象与利用的框架。

SAGA progressively transforms interaction trajectories into episodic descriptions, reusable procedures, and principles with explicit applicability conditions, while maintaining links to execution evidence. Retrieved principles are instantiated into task-specific guidance and used to refine candidate actions through corrective feedback and resampling.

SAGA 将交互轨迹逐步转化为情景描述、可重用程序以及具有明确适用条件的原则，同时保持与执行证据的联系。检索到的原则被实例化为特定任务的指导，并通过纠正性反馈和重采样用于优化候选行动。

This creates an execution—abstraction feedback loop, where accumulated knowledge guides future interactions and new experiences continuously update hierarchical memory. Experiments on ScienceWorld and ALFWorld demonstrate improved task performance, with ablation studies highlighting the importance of contextual instantiation and action regulation for leveraging principle-level knowledge.

这创造了一个“执行-抽象”反馈循环，其中积累的知识指导未来的交互，而新的经验不断更新分层记忆。在 ScienceWorld 和 ALFWorld 上的实验证明了任务性能的提升，消融研究强调了上下文实例化和行动调节对于利用原则级知识的重要性。