arXiv 2602.140352026-02cs.AI

FloCA: 向忠实且逻辑一致的流程图推理迈进

FloCA: Towards Faithful and Logically Consistent Flowchart Reasoning

MoE KLINNS Lab, Xi’an Jiaotong University, Xi’an 710049, P. R. China（西安交通大学）；China Mobile Group Shaanxi Co., Ltd.（中国移动集团陕西公司）

Jinzi Zou, Bolin Wang, Shuo Zhang, Nuo Xu, Junzhou Zhao

备注：: Accepted to the 2026 Conference on Empirical Methods in Natural Language Processing (EMNLP 2026)

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

FloCA通过结合LLM和外部工具实现流程图推理，解决现有方法在流程图拓扑表示和幻觉问题上的不足，提升对话系统的准确性和一致性。

摘要：

面向流程图的对话（FOD）系统旨在通过遵循特定领域的流程图来引导用户进行多轮决策或操作流程，以实现任务目标。在本文中，我们将FOD中的流程图推理正式化为在每次对话回合中将用户输入接地到流程图节点，同时确保节点转换与正确的流程图路径一致。尽管近年来在任务导向对话系统中LLM取得了进展，但将其应用于FOD仍面临两个限制：（1）LLM缺乏明确表示和推理流程图拓扑的机制；（2）它们容易产生幻觉，导致不忠实的流程图推理。为了解决这些限制，我们提出了FloCA，一个零样本的面向流程图的对话代理。FloCA使用LLM进行意图理解和响应生成，同时将流程图推理委托给一个外部工具，该工具执行受拓扑约束的图执行，确保在对话回合中节点转换的忠实和逻辑一致性。我们进一步引入了一个评估框架，包含基于LLM的用户模拟器和五个新的指标，涵盖推理准确性和交互效率。在FLODIAL和PFDial数据集上的大量实验突显了现有基于LLM的方法的瓶颈，并展示了FloCA的优越性。我们的代码可在https://github.com/Jinzi-Zou/FloCA-flowchart-reasoning上获得。

英文摘要：

Flowchart-oriented dialogue (FOD) systems aim to guide users through multi-turn decision-making or operational procedures by following a domain-specific flowchart to achieve a task goal. In this work, we formalize flowchart reasoning in FOD as grounding user input to flowchart nodes at each dialogue turn while ensuring node transition is consistent with the correct flowchart path. Despite recent advances of LLMs in task-oriented dialogue systems, adapting them to FOD still faces two limitations: (1) LLMs lack an explicit mechanism to represent and reason over flowchart topology, and (2) they are prone to hallucinations, leading to unfaithful flowchart reasoning. To address these limitations, we propose FloCA, a zero-shot flowchart-oriented conversational agent. FloCA uses an LLM for intent understanding and response generation, while delegating flowchart reasoning to an external tool that performs topology-constrained graph execution, ensuring faithful and logically consistent node transitions across dialogue turns. We further introduce an evaluation framework with an LLM-based user simulator and five new metrics covering reasoning accuracy and interaction efficiency. Extensive experiments on FLODIAL and PFDial datasets highlight the bottlenecks of existing LLM/VLM-based methods and demonstrate the superiority of FloCA. The code and dataset are publicly available at https://github.com/Jinzi-Zou/FloCA-flowchart-reasoning.

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。