InFi-Check: 可解释且细粒度的大型语言模型事实核查
InFi-Check: Interpretable and Fine-Grained Fact-Checking of LLMs
AI总结：
InFi-Check通过可控数据合成和细粒度事实核查框架，提升大型语言模型事实性评估的可解释性和准确性。
摘要：
大型语言模型（LLMs）经常产生幻觉，但现有的大多数事实核查方法将事实性评估视为二元分类问题，提供有限的可解释性，并未能捕捉到细粒度的错误类型。在本文中，我们介绍了InFi-Check，一种用于可解释且细粒度核查LLM输出的框架。具体来说，我们首先提出了一条受控的数据合成管道，生成高质量的数据，其特征包括显式的证据、细粒度的错误类型标签、说明和修正。基于此，我们进一步构建了大规模训练数据和一个人工验证的基准InFi-Check-FG，用于LLM输出的细粒度事实核查。在这些高质量训练数据的基础上，我们进一步提出了InFi-Checker，它可以联合提供支持证据、分类细粒度的错误类型，并生成说明和修正。实验表明，InFi-Checker在InFi-Check-FG上实现了最先进的性能，并在各种下游任务中表现出强大的泛化能力，显著提高了事实性评估的实用性和可信度。
英文摘要：
Large language models (LLMs) often hallucinate, yet most existing fact-checking methods treat factuality evaluation as a binary classification problem, offering limited interpretability and failing to capture fine-grained error types. In this paper, we introduce InFi-Check, a framework for interpretable and fine-grained fact-checking of LLM outputs. Specifically, we first propose a controlled data synthesis pipeline that generates high-quality data featuring explicit evidence, fine-grained error type labels, justifications, and corrections. Based on this, we further construct large-scale training data and a manually verified benchmark InFi-Check-FG for fine-grained fact-checking of LLM outputs. Building on these high-quality training data, we further propose InFi-Checker, which can jointly provide supporting evidence, classify fine-grained error types, and produce justifications along with corrections. Experiments show that InFi-Checker achieves state-of-the-art performance on InFi-Check-FG and strong generalization across various downstream tasks, significantly improving the utility and trustworthiness of factuality evaluation.
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。
