Towards Comprehensive Stage-wise Benchmarking of Large Language Models in Fact-Checking

TL;DR

This paper introduces FactArena, a comprehensive evaluation framework for large language models that assesses their performance across the entire fact-checking process, revealing systematic weaknesses and guiding future improvements.

Contribution

The paper presents a novel, fully automated, stage-wise benchmarking framework that evaluates LLMs on claim decomposition, evidence retrieval, and justification, addressing limitations of existing single-stage benchmarks.

Findings

Significant discrepancies between claim verification accuracy and end-to-end fact-checking performance.

FactArena provides stable, interpretable rankings across 16 state-of-the-art LLMs.

Adaptive claim evolution probes models' factual robustness beyond fixed datasets.

Abstract

Large Language Models (LLMs) are increasingly deployed in real-world fact-checking systems, yet existing evaluations focus predominantly on claim verification and overlook the broader fact-checking workflow, including claim extraction and evidence retrieval. This narrow focus prevents current benchmarks from revealing systematic reasoning failures, factual blind spots, and robustness limitations of modern LLMs. To bridge this gap, we present FactArena, a fully automated arena-style evaluation framework that conducts comprehensive, stage-wise benchmarking of LLMs across the complete fact-checking pipeline. FactArena integrates three key components: (i) an LLM-driven fact-checking process that standardizes claim decomposition, evidence retrieval via tool-augmented interactions, and justification-based verdict prediction; (ii) an arena-styled judgment mechanism guided by consolidated reference guidelines to ensure unbiased and consistent pairwise comparisons across heterogeneous judge agents; and (iii) an arena-driven claim-evolution module that adaptively generates more challenging and semantically controlled claims to probe LLMs' factual robustness beyond fixed seed data. Across 16 state-of-the-art LLMs spanning seven model families, FactArena produces stable and interpretable rankings. Our analyses further reveal significant discrepancies between static claim-verification accuracy and end-to-end fact-checking competence, highlighting the necessity of holistic evaluation. The proposed framework offers a scalable and trustworthy paradigm for diagnosing LLMs' factual reasoning, guiding future model development, and advancing the reliable deployment of LLMs in safety-critical fact-checking applications.