用大模型从对话记录自动提炼策略，让非协作对话智能体更高效构建
METRO: Towards Strategy Induction from Expert Dialogue Transcripts for Non-collaborative Dialogues
- 通过语言模型从原始对话中提取策略动作与规划逻辑
- 在两个基准上平均超越现有方法9%-10%的性能
- 适合想低成本构建对话智能体的研究者与开发者
传统非协作对话智能体的开发依赖人工编写专家策略，难以扩展。本文提出METRO，利用大语言模型从原始对话转录文本中自主推导策略动作与规划逻辑。METRO将专家知识形式化为策略森林（Strategy Forest），一种分层结构，既能捕捉短期回应（节点），又能体现长期战略预见性（分支）。在两个基准上的实验表明，METRO性能显著优于现有方法，平均提升9%-10%。进一步分析揭示其成功源于策略行为多样性与前瞻性，并验证了跨任务强迁移能力。该方法为低成本、可扩展地构建非协作对话智能体提供了新路径。代码已开源：https://github.com/Humphrey-0125/METRO。
原文摘要 · Abstract (English)
Developing non-collaborative dialogue agents traditionally requires the manual, unscalable codification of expert strategies. We propose \ours, a method that leverages large language models to autonomously induce both strategy actions and planning logic directly from raw transcripts. METRO formalizes expert knowledge into a Strategy Forest, a hierarchical structure that captures both short-term responses (nodes) and long-term strategic foresight (branches). Experimental results across two benchmarks show that METRO demonstrates promising performance, outperforming existing methods by an average of 9%-10%. Our further analysis not only reveals the success behind METRO (strategic behavioral diversity and foresight), but also demonstrates its robust cross-task transferability. This offers new insights into building non-collaborative agents in a cost-effective and scalable way. Our code is available at https://github.com/Humphrey-0125/METRO.
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。