大语言模型智能体行为分析框架提出：从数据驱动发现到实验验证
Behaviour analysis for large language model-driven agents
AI 导读
研究者提出一个分析大语言模型智能体行为的通用框架，将数据驱动的行为发现与分析和实验研究相连接。框架基于记录活动建立行为类别，保留行为在轨迹中的位置并形成可解释假设，通过统计评估、held-out 上下文测试和提示词操控实验检验行为与结果的关系。以迭代自我反思为案例展示了发现、依赖学习与提示词对比的相互印证，方法支持不同长度、事件顺序和智能体角色的轨迹。
正文 · 原文
Abstract:Understanding language-agent behaviour requires learning how actions are organised across a trajectory and how this organisation relates to context and outcomes. We propose a general framework that connects data-driven behavioural discovery with analysis and experimental investigation. Categories developed from recorded activity define what is measured. Analyses preserve the behaviours' positions in the process and express their relationships as interpretable hypotheses. Statistical assessment and tests in held-out contexts examine the strength and scope of the evidence. Experiments, including prompt steering, then test how behaviour and outcomes change under deliberately varied conditions. A study of iterative self-reflection illustrates how discovery, dependency learning and prompt comparisons can inform one another. We develop this approach for trajectories of varying length, event order and agent roles, and identify advances in coding, analysis and experimentation needed to support its wider use.
| Comments: | Under review by Nature Machine Intelligence |
| Subjects: | Artificial Intelligence (cs.AI); Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.32087 [cs.AI] |
| (or arXiv:2610.32087v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.32087 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Tianqiang Yan [view email]
[v1]
Fri, 23 Oct 2026 02:48:45 UTC (464 KB)
来源：arXiv cs.AI · arxiv.org