多模态声明提取用于事实核查
Multimodal Claim Extraction for Fact-Checking
AI总结：
本文提出首个多模态社交媒体声明提取基准，评估了最先进的多模态大语言模型，在语义对齐、忠实度和去上下文化框架下发现基线模型难以建模修辞意图和上下文线索，引入意图感知框架MICE提升关键意图场景性能。
摘要：
自动化事实核查（AFC）依赖声明提取作为第一步，但现有方法忽视了当今虚假信息的多模态特性。社交媒体帖子常结合简短非正式文本与图片如表情包、截图和照片，挑战不同于纯文本声明提取和已研究的多模态任务如图像描述或视觉问答。本文提出首个多模态社交媒体声明提取基准，包含文本和一张或多张图片的帖子，标注有来自现实世界事实核查者的黄金标准声明。我们评估了最先进的多模态大语言模型（MLLMs）在三部分评估框架（语义对齐、忠实度和去上下文化）下，发现基线MLLMs难以建模修辞意图和上下文线索。为解决此问题，我们引入MICE，一个意图感知框架，在意图关键场景中表现出改进。
英文摘要：
Automated Fact-Checking (AFC) relies on claim extraction as a first step, yet existing methods largely overlook the multimodal nature of today's misinformation. Social media posts often combine short, informal text with images such as memes, screenshots, and photos, creating challenges that differ from both text-only claim extraction and well-studied multimodal tasks like image captioning or visual question answering. In this work, we present the first benchmark for multimodal claim extraction from social media, consisting of posts containing text and one or more images, annotated with gold-standard claims derived from real-world fact-checkers. We evaluate state-of-the-art multimodal LLMs (MLLMs) under a three-part evaluation framework (semantic alignment, faithfulness, and decontextualization) and find that baseline MLLMs struggle to model rhetorical intent and contextual cues. To address this, we introduce MICE, an intent-aware framework which shows improvements in intent-critical cases.
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。
