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 2511.07086cs.AI
收藏

LLM驱动流程以促进可解释人工智能

LLM Driven Processes to Foster Explainable AI

University of Applied Sciences Ruhr West(鲁尔西部应用科学大学)

Marcel Pehlke, Marc Jansen

AI总结:

本文提出一种模块化、可解释的LLM智能体流水线,通过结合Vester敏感度模型、标准式博弈和序贯博弈,并配合确定性分析器,在物流案例中验证其能以透明可审计的步骤模拟专家决策流程。

摘要:

我们提出了一种模块化、可解释的LLM智能体流水线,用于决策支持,将推理过程外部化为可审计的中间产物。该系统实例化了三个框架:Vester敏感度模型(因子集、带符号影响矩阵、系统角色、反馈回路);标准式博弈(策略、收益矩阵、均衡);以及序贯博弈(角色条件化智能体、树构建、逆向归纳),且每一步的模块均可替换。LLM组件(默认:GPT-5)与用于均衡和基于矩阵的角色分类的确定性分析器配对,从而产生可追溯的中间结果,而非不透明的输出。在一个真实物流案例中(100次运行),26个因子上的平均因子对齐度与人类基线相比为55.5%,在运输核心子集上为62.9%;匹配项上的角色一致性为57%。一个使用八项标准评分表(满分100)的LLM评判者对运行的评分与重建的人类基线相当。因此,可配置的LLM流水线能够以透明、可检查的步骤模拟专家工作流程。

英文摘要:

We present a modular, explainable LLM-agent pipeline for decision support that externalizes reasoning into auditable artifacts. The system instantiates three frameworks: Vester's Sensitivity Model (factor set, signed impact matrix, systemic roles, feedback loops); normal-form games (strategies, payoff matrix, equilibria); and sequential games (role-conditioned agents, tree construction, backward induction), with swappable modules at every step. LLM components (default: GPT-5) are paired with deterministic analyzers for equilibria and matrix-based role classification, yielding traceable intermediates rather than opaque outputs. In a real-world logistics case (100 runs), mean factor alignment with a human baseline was 55.5\% over 26 factors and 62.9\% on the transport-core subset; role agreement over matches was 57\%. An LLM judge using an eight-criterion rubric (max 100) scored runs on par with a reconstructed human baseline. Configurable LLM pipelines can thus mimic expert workflows with transparent, inspectable steps.

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。

↑