arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2605.09192cs.AI
收藏

证据与计划:用于技能蒸馏的在线轨迹验证

Evidence Over Plans: Online Trajectory Verification for Skill Distillation

Rutgers University(罗杰斯大学)

Yang Zhou, Zihan Dong, Zhenting Wang, Can Jin, Shiyu Zhao, Bangwei Guo, Difei Gu, Linjun Zhang, Mu Zhou, Dimitris N. Metaxas

AI总结:

本文提出了一种基于轨迹的后验蒸馏指数(PDI)来评估技能与任务环境证据的契合度,通过SPARK框架实现环境验证轨迹的生成,从而提升技能的效率和可迁移性。

摘要:

代理技能可以通过使用人类编写的程序性文档显著提高任务成功率,但其质量在没有环境基础验证的情况下难以评估。现有的技能生成方法严重依赖于偏好日志而不是直接的环境交互,通常产生微不足道甚至退化的收益。我们发现这是一个根本的时间瓶颈:稳健的技能应基于后验,从经验环境交互中蒸馏,而不是先验计划。在本研究中,我们引入了后验蒸馏指数(PDI),这是一个轨迹级指标,量化了蒸馏技能与任务-环境证据的契合程度。为了操作化PDI,我们提出了SPARK(用于自主可运行任务和技能生成的结构化流程),以保留任务执行证据以实现全面的轨迹级分析。SPARK生成用于计算PDI的环境验证轨迹,并将其用作在线诊断和干预信号,以确保后验技能的形成。在86个可运行任务上,SPARK生成的技能始终优于无技能基线,并在学生模型上优于人工编写技能(推理成本比教师模型低高达1000倍)。这些发现表明,PDI引导的蒸馏产生了高效且可迁移的技能,这些技能基于任务-环境交互。我们发布代码在https://github.com/EtaYang10th/spark-skills。

英文摘要:

Agent skills can remarkably improve task success rates by using human-written procedural documents, but their quality is difficult to assess without environment-grounded verification. Existing skill generation methods heavily rely on preference logs rather than direct environment interaction, often yielding negligible or even degraded gains. We identify that it is a fundamental timing bottleneck: robust skills should be posterior-based, distilled from empirical environment interaction rather than prior plans. In this study, we introduce the Posterior Distillation Index (PDI), a trajectory-level metric that quantifies how well a distilled skill is grounded in the task-environment evidence. To operationalize PDI, we present SPARK (Structured Pipelines for Autonomous Runnable tasKs and sKill generation) for preserving task execution evidence towards full trajectory-level analysis. SPARK generates environment-verified trajectories used to compute PDI, and it applies PDI as an online diagnostic and intervention signal to ensure posterior skill formation. Across 86 runnable tasks, SPARK-generated skills consistently surpass no-skill baselines and outperform human-written skills on student models (inference cost up to 1,000x cheaper than teacher models). These findings show that PDI-guided distillation produces efficient and transferable skills grounded in the task-environment interaction. We release our code at https://github.com/EtaYang10th/spark-skills .

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑