MemoHarness：从经验中学习的智能体控制层

中文摘要 AI 辅助

智能体控制层是通过管理上下文、工具、编排、内存、解码和输出处理，将基础语言模型转变为可执行智能体的外部控制层。虽然控制层设计对智能体行为有重大影响，但多数自动改进方法针对提示、管道或工作流程等更窄的工件进行优化，且部署的智能体通常对所有情况都重用单一全局控制层。我们引入了MemoHarness，这是一个从自身执行中学习的自适应控制层优化框架。MemoHarness将控制层分解为六个可编辑的控制维度，在双层经验库中存储每个案例的诊断结果和提炼的全局模式，并利用检索到的经验将学习到的控制层应用于每个测试用例，无需测试时标签反馈或额外搜索。在我们对shell智能体、代码生成和分析推理基准的评估中，MemoHarness优于我们比较的固定控制层，并显示出向未见套件和基础模型的选择性迁移。当大部分检索到的经验可缓存时，其额外上下文也能保持成本竞争力。这些结果证明，执行经验是构建比单一静态配置更具适应性的智能体控制层的实用基础，而关于统计稳健性和组件归因的更广泛主张留待未来工作。

英文摘要

An agent harness is the external control layer that turns a base LLM into an executable agent by managing context, tools, orchestration, memory, decoding, and output handling. While harness design strongly affects agent behavior, most automatic improvement methods optimize narrower artifacts such as prompts, pipelines, or workflows, and deployed agents usually reuse a single global harness for all cases. We introduce MemoHarness, an adaptive harness optimization framework that learns from its own executions. MemoHarness decomposes the harness into six editable control dimensions, stores per-case diagnoses and distilled global patterns in a dual-layer experience bank, and adapts the learned harness to each test case using retrieved experience without test-time labels, feedback, or additional search. In our evaluation across shell-agent, code-generation, and analytical-reasoning benchmarks, MemoHarness improves over the fixed harnesses we compare against and shows selective transfer to unseen suites and base models. Its additional context can also remain cost-competitive when much of the retrieved experience is cacheable. These results provide evidence that execution experience is a practical substrate for building agent harnesses that are more adaptive than a single static configuration, while leaving broader claims about statistical robustness and component attribution to future work.

发表机构

• University of Notre Dame（圣母大学）
• LMU Munich（慕尼黑大学）
• University of Southern California（南加州大学）

机构由 AI 辅助整理，请以论文原文为准。