RUPA：面向LLM代理的关系不确定性传播框架
原标题：Paper page - From Sequence to Structure: Relational Uncertainty Propagation for LLM Agents
AI 摘要
RUPA是一个针对LLM代理的轨迹级不确定性量化框架，它将执行历史建模为有向图，通过传播不确定性来捕捉错误在长轨迹中的累积和转移。在τ-2、Terminal-Bench-2和GAIA基准测试中，使用6个开源LLM进行的实验表明，RUPA在准确性、早期失败检测和不确定性引导执行方面优于现有方法。该研究强调了显式建模关系依赖对于长周期LLM代理可靠不确定性量化的重要性。
正文节选
From Sequence to Structure: Relational Uncertainty Propagation for LLM Agents Abstract RUPA models agent execution as a dependency graph to propagate uncertainty across long trajectories, improving failure detection and confidence estimation for LLM agents. Reliable uncertainty quantification (UQ) is essential for deploying large language model (LLM) agents in complex interactive environments. Existing UQ methods largely rely on local signals, such as token probabilities, predictive entropy, or