通过递归记忆进化提升长时任务智能体的自改进能力
Recursive Experiential-Working Memory Evolution for Long-Horizon Agent Harnesses
- 设计体验-工作记忆耦合机制，按当前任务需求选择技能而非依赖完整历史
- 在4个基准上35/37组合提升成功率，最长任务提升达32.2分
- 适合需要持续学习与长程规划的AI代理系统研究者
长时任务中递归自改进（RSI）仍具挑战，因历史信息膨胀导致任务状态模糊和技能调用错配。本文提出Recuris，一种递归体验-工作记忆架构，其中工作记忆追踪任务进展并指导从经验记忆中选择技能，使技能使用基于当前需求而非完整历史。该耦合机制将执行过程转化为结构化证据，可定位失败至特定记忆组件。一个固定元智能体利用这些证据，生成受验证约束的技能记忆更新，重塑执行并产生新证据，形成闭环递归记忆演化。在四个长时基准和十种模型上，Recuris在37组完成实验中提升了35组任务成功率：在tau-bench上使GPT-5.6 Sol提升+17.8分、Claude Opus 5提升+15.6分（达87.9%），在SkillFlow上对Qwen3.6-27B/35B分别提升+16.6/+13.5分；随着交互时域增长，优势扩大至+32.2分，常见长时失败减少高达80%。结果表明递归演化记忆是可扩展的RSI基础，使智能体持续将经验转化为更有效的长时行为。
原文摘要 · Abstract (English)
Recursive self-improvement (RSI) remains hard in long-horizon tasks, where growing histories obscure the task state and misalign skill invocation. We introduce Recuris, a recursive Experiential-Working Memory architecture for long-horizon agent harnesses, in which Working Memory tracks task progress and guides skill selection from Experiential Memory, grounding skill use in current needs rather than the full history. This coupling also turns execution into structured evidence that localizes failures to specific memory components. Across tasks, a fixed Meta-Agent turns that evidence into localized, validation-gated updates to Skill Memory that reshape execution and yield new evidence, forming a bounded recursive memory-evolution loop. Across four long-horizon benchmarks and ten models, Recuris improves task success in 35 of the 37 completed model-benchmark pairs, carrying frontier models to SOTA-level task success: on tau-bench it adds +17.8 points to GPT-5.6 Sol and +15.6 to Claude Opus 5, taking Opus 5 to 87.9%, and +16.6/+13.5 points on Qwen3.6-27B/35B on SkillFlow. The advantage widens as the interaction horizon grows, to +32.2 points on the longest tasks, and common long-horizon failures fall by up to 80%. These results position recursively evolving memory as a scalable foundation for RSI, enabling agents to continuously transform accumulated experience into increasingly effective long-horizon behavior. Code: https://github.com/Gen-Verse/Recuris
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。