让网页导航智能体学会反思，避免重复犯错。
Reflection-Based Memory For Web navigation Agents
- 通过自我反思记录成功与失败经验
- 整体性能提升11分，旧错误任务提升29分
- 适合需要持续学习的智能交互系统
网页导航智能体虽有显著进展，但当前系统缺乏对过往经历的记忆，导致重复错误且无法从互动中学习。本文提出反射增强规划（ReAP）方法，利用自我反思机制整合成功与失败的历史经验。实验表明，该方法在整体性能上比基线提升11个百分点，在曾失败的任务上提升达29个百分点。结果证明，反思能力可迁移至不同网页导航任务，显著提升智能体的自适应与学习能力。
原文摘要 · Abstract (English)
Web navigation agents have made significant progress, yet current systems operate with no memory of past experiences -- leading to repeated mistakes and an inability to learn from previous interactions. We introduce Reflection-Augment Planning (ReAP), a web navigation system to leverage both successful and failed past experiences using self-reflections. Our method improves baseline results by 11 points overall and 29 points on previously failed tasks. These findings demonstrate that reflections can transfer to different web navigation tasks.
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。