InFi-Check: Interpretable and Fine-Grained Fact-Checking of LLMs

Yuzhuo Bai, Shuzheng Si, Kangyang Luo, Qingyi Wang, Wenhao Li +3 more
1/10/2026
cs.CLcs.AI

No official code found. Best implementation: Phosphor-Bai/InFi-Check (12 stars, last commit 2026-01-20).

Abstract

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.

View on arXivView PDF

Code Implementations(1)

Phosphor-Bai/InFi-CheckCommunity68%

Source code for "InFi-Check: Interpretable and Fine-Grained Fact-Checking of LLMs"

120PythonJan 10, 20269 months ago

Similar Papers

Towards Automated Fact-Checking of Real-World Claims: Exploring Task Formulation and Assessment with LLMs

86% match2025

Hallucination to Truth: A Review of Fact-Checking and Factuality Evaluation in Large Language Models

86% match2025

FActBench: A Benchmark for Fine-grained Automatic Evaluation of LLM-Generated Text in the Medical Domain

86% match2025

FACT-AUDIT: An Adaptive Multi-Agent Framework for Dynamic Fact-Checking Evaluation of Large Language Models

86% match1 repo192025

Evidence-backed Fact Checking using RAG and Few-Shot In-Context Learning with LLMs

85% match1 repo72024

Multi-Modal Fact-Verification Framework for Reducing Hallucinations in Large Language Models

85% match2025

Cite this paper

@article{bai2026inficheck,
  title  = {InFi-Check: Interpretable and Fine-Grained Fact-Checking of LLMs},
  author = {Yuzhuo Bai and Shuzheng Si and Kangyang Luo and Qingyi Wang and Wenhao Li and Gang Chen and Fanchao Qi and Maosong Sun},
  year   = {2026},
  eprint = {2601.06666},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2601.06666v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2601.06666.svg)](https://codeofpaper.com/paper/2601.06666)

Discussion