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 2606.24151cs.CLcs.AI
收藏

Metis: 连接文本与代码记忆以实现自我进化智能体

Metis: Bridging Text and Code Memory for Self-Evolving Agents

The Chinese University of Hong Kong(香港中文大学);Huawei(华为);Wuhan University(武汉大学)

Zijie Dai, Siuhin He, Hui Li, Qihui Zhou, Jiajun Li, Mingcong Song, Guoping Long, Hongjie Si, Xin Yao, Lin Zhang, James Cheng, Xiao Yan

AI总结:

提出Metis系统,通过分层双表示记忆(文本与代码)平衡构建成本、执行效率和可迁移性,在AppWorld上任务准确率提升20.6%,执行成本降低22.8%。

摘要:

自我进化智能体通过从过去执行中提炼经验并在未来任务中重用这些经验来随时间改进。现有系统将此类经验表示为注入到智能体上下文中的自然语言文本或作为可调用工具暴露的代码。然而,这些表示之间的选择通常在设计时做出,而非从经验本身的特性中推导,导致它们之间的权衡关系尚不明确。我们首次进行了控制研究,在相同的经验集上隔离了文本记忆和代码记忆。结果表明,两种形式在构建成本、执行效率和可迁移性方面表现出互补的权衡,使得单独任何一种表示都不足够。受这些发现启发,我们提出了Metis,一个基于分层双表示记忆的自我进化智能体系统。Metis将文本经验组织为执行计划、环境事实和常见陷阱,并选择性地将重复出现的计划结晶为经过验证的可调用工具。这种设计结合了文本记忆的广泛适用性和代码记忆的执行效率,同时仅在重复使用证明合理时才产生工具生成成本。我们在交互式智能体的挑战性基准AppWorld上评估了Metis。结果表明,与ReAct相比,Metis将任务准确率提高了20.6%,同时将执行成本降低了22.8%。与代表性的自我进化智能体系统相比,Metis在准确性、执行效率和记忆构建成本之间始终实现了更好的平衡。

英文摘要:

Self-evolving agents improve over time by distilling experience from past executions and reusing it in future tasks. Existing systems represent such experience either as natural-language text injected into the agent context or as code exposed as callable tools. However, the choice between these representations is typically made at design time rather than derived from the characteristics of the experience itself, leaving the trade-offs between them poorly understood. We present the first controlled study that isolates text memory and code memory over an identical set of experiences. Our results show that the two forms exhibit complementary trade-offs in construction cost, execution efficiency, and transferability, such that neither representation alone is sufficient. Guided by these findings, we propose Metis, a self-evolving agent system built on a hierarchical dual-representation memory. Metis organizes textual experience into execution plans, environment facts, and common pitfalls, and selectively crystallizes recurring plans into validated callable tools. This design combines the broad applicability of text memory with the execution efficiency of code memory while incurring tool-generation cost only when justified by repeated reuse. We evaluate Metis on AppWorld, a challenging benchmark for interactive agents. The results show that Metis improves task accuracy by up to 20.6% over ReAct while reducing execution cost by up to 22.8%. Compared with representative self-evolving agent systems, Metis consistently achieves a better balance between accuracy, execution efficiency, and memory-construction cost.

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。

↑