arXiv：cs.AI· Jaeyoon Jung, Yejun Yoon, Kunwoo Park·· 4 小时前AI 评分39
AMuFC：自适应多模态事实核查框架，按需使用视觉证据
Is a Picture Worth a Thousand Words? Adaptive Multimodal Fact-Checking with Visual Evidence Necessity
AI 导读
研究挑战了"引入视觉证据必然提升核查准确率"的假设，发现不加区分地使用视觉证据反而会降低准确率。为此提出模块化事实核查框架 AMuFC，用两个分工协作的视觉语言模型实现视觉证据的自适应使用。在包括本研究提出的 WebFC 在内的三个数据集上，实验验证了自适应使用视觉证据的有效性。
正文
Abstract:Automated fact-checking is a crucial task that supports a responsible information ecosystem. While recent research has progressed from text-only to multimodal fact-checking, a prevailing assumption is that incorporating visual evidence universally improves verification accuracy. In this work, we challenge this assumption and show that the indiscriminate use of visual evidence can reduce accuracy. Building on this finding, we propose AMuFC, a modular fact-checking framework that employs two collaborative vision-language models with distinct roles to enable the adaptive use of visual evidence. Experimental results on three datasets, including WebFC, introduced in this study, demonstrate the effectiveness of adaptive visual evidence use in fact-checking.
| Comments: | AACL-IJCNLP 2026 |
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | arXiv:2604.04692 [cs.CL] |
| (or arXiv:2604.04692v3 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2604.04692 arXiv-issued DOI via DataCite |
Submission history
From: Kunwoo Park [view email]
[v1]
Mon, 6 Apr 2026 14:01:38 UTC (5,738 KB)
[v2]
Wed, 13 May 2026 06:23:13 UTC (5,742 KB)
[v3]
Fri, 2 Oct 2026 01:16:11 UTC (5,744 KB)
来源：arXiv：cs.AI · arxiv.org