通过文本嵌入分析代理行为演变，精准追踪其敏感数据获取倾向变化。
Tracking the Behavioral Trajectories of Adapting Agents
- 用文本嵌入差分构建行为特征向量，量化代理行为改变方向。
- 在68个标注样本上实现91.2%符号分类准确率和0.82的秩相关性。
- 可部署于代理间信任协议，实现对技能更新的可信评估。
技能文件、记忆文件和行为配置文件等文本文件在定义现代智能体行为中起核心作用。这些文件可能随人工或智能体自身修改而演化，直接影响其后续交互行为。本文提出一种方法，将智能体‘特质’定义为文本嵌入模型中的方向。通过在标注的‘前-后’技能文件差异上训练线性模型，学习出特质向量，并通过投影任意技能修改的嵌入差分来评分。在68个标注的技能差异对上评估敏感数据获取倾向时，该方法在留一交叉验证下达到91.2%的符号分类准确率和ρ=0.82的斯皮尔曼秩相关系数。我们将此特质评估集成到一个代理间协议中，使一个代理可通过可信中介评估另一个代理的技能文件更新。
原文摘要 · Abstract (English)
Text files such as skill files, memory files, and behavioral configuration files play a central role in defining how modern agents act. Through edits by humans or the agents themselves, these files may evolve over time, directly steering the agent's behavior in future interactions. We present a methodology and framework for measuring agent $traits$ by defining traits as directions in the embedding space of a text embedding model. We train a linear model on labeled "before" versus "after" skill file diffs to learn a trait vector, then score arbitrary skill edits by projecting their embedding diffs onto this vector. Evaluated on 68 labeled skill diff pairs for the trait of propensity to seek sensitive data, our method achieves 91.2% sign classification accuracy and a Spearman rank correlation of $ρ= 0.82$ under leave-one-out cross-validation. We build this trait evaluation into a broader agent-to-agent protocol that enables one agent to evaluate another's skill file updates through a trusted intermediary.
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。