让大模型学会从行为轨迹中自动提炼多层级技能，提升任务完成效率。
Trace2Tower: Transition-Aware EigenTrace Induction of Multi-Level Skills for LLM Agents
- 基于语义兼容性与动态转移构建统一行为图，捕捉动作间的时序关系。
- 在ALFWorld上仅用10.35步达成87.31%成功率，错误动作仅0.26个。
- 适合研究智能体技能抽象、强化学习与复杂任务规划的开发者。
大型语言模型代理越来越依赖执行轨迹来掌握复杂的交互任务。然而，当前方法受限于浅层轨迹检索和平面化技能总结，从根本上忽略了行为的时间依赖性和结果导向的拓扑结构。我们提出Trace2Tower，一种面向转移的EigenTrace框架，将原始轨迹提炼为稳健的技能层级结构。该框架将步骤级交互抽象为标准事件，构建由语义兼容性、转移动态和结果证据共同驱动的统一图结构。通过一种新颖的对比谱分解，它分离出稳定且成功对齐的行为模式，同时严格抑制失败倾向的捷径。这些模式自然形成包含动作模板、程序化流程和高层任务策略的动态技能塔，并通过验证器引导的反馈持续优化。在ALFWorld上，Trace2Tower实现了87.31%的成功率，仅需10.35步和0.26次无效操作；在WebShop上达到50.67%的精确成功率。在两个基准测试中，其任务掌握能力和上下文高效经验复用均显著优于现有基线。
原文摘要 · Abstract (English)
Large language model agents increasingly rely on execution traces to master complex interactive tasks. However, current paradigms are bottlenecked by shallow trajectory retrieval and flat skill summarization, fundamentally ignoring the temporal dependencies and outcome-conditioned topology of agent behavior. We introduce Trace2Tower, a transition-aware EigenTrace framework that distills raw trajectories into a robust skill hierarchy. Trace2Tower abstracts step-level interactions into canonical events, constructing a unified graph governed by semantic compatibility, transition dynamics, and outcome evidence. Through a novel contrastive spectral decomposition, it isolates stable, success-aligned behavioral modes while rigorously suppressing failure-prone shortcuts. These modes organically populate a dynamic skill tower of action templates, procedural routines, and overarching task strategies, continuously refined via verifier-guided feedback. On ALFWorld, Trace2Tower achieves 87.31% success requiring only 10.35 steps and 0.26 invalid actions; on WebShop, it reaches 50.67% exact success. Across both benchmarks, Trace2Tower significantly outperforms existing baselines in task mastery and context-efficient experience reuse.
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。