dorsal/arxiv
View SchemaInFi-Check: Interpretable and Fine-Grained Fact-Checking of LLMs
| Authors | Yuzhuo Bai, Shuzheng Si, Kangyang Luo, Qingyi Wang, Wenhao Li, Gang Chen, Fanchao Qi, Maosong Sun |
|---|---|
| Categories | |
| ArXiv ID | 2601.06666vv1 |
| URL | https://arxiv.org/abs/2601.06666 |
| License | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ |
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.
{
"annotation_id": "3f13093d-76c0-454d-aee9-9fb81bdd9279",
"date_created": "2026-02-17T05:53:08.609000Z",
"date_modified": "2026-02-17T05:53:08.609000Z",
"file_hash": "2b7b20a4fcc885e619511aff9fa673316519f93b0956ac3aa55e80be5128121b",
"private": false,
"record": {
"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.",
"arxiv_id": "2601.06666",
"authors": [
"Yuzhuo Bai",
"Shuzheng Si",
"Kangyang Luo",
"Qingyi Wang",
"Wenhao Li",
"Gang Chen",
"Fanchao Qi",
"Maosong Sun"
],
"categories": [
"cs.CL",
"cs.AI"
],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"title": "InFi-Check: Interpretable and Fine-Grained Fact-Checking of LLMs",
"url": "https://arxiv.org/abs/2601.06666",
"version": "v1"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "ed320fbd-4ee7-45f3-b4b3-fbe086a5529a",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}