构建可审计的多模态假信息检测数据集，提升模型对图文证据的可信关联能力。
RW-Post: Auditable Evidence-Grounded Multimodal Fact-Checking in the Wild
- 基于真实社交媒体帖子构建带推理链与证据链接的多模态事实核查数据集
- 现有模型在证据定位上仍有明显不足，证据受限评估显著提升准确率与可信度
- 适合研究多模态事实核查、视觉证据理解及可解释性验证的学者与工程师
多模态虚假信息越来越多利用视觉说服力，通过改编或篡改图片强化误导性文本。我们提出 RW-Post，一个面向真实场景的图文对齐事实核查基准，具备可审计标注：每个实例均关联原始社交媒体帖子、推理轨迹以及通过 LLM 辅助提取与审核流程从人工事实核查文章中获取的明确证据项。RW-Post 支持在封闭知识、证据受限和开放网络三种模式下的可控评估，可系统诊断视觉锚定与证据使用能力。我们提供 AgentFact 作为参考验证基线，并在统一协议下评测多个开源多模态大模型。实验表明当前模型在忠实证据锚定方面仍有巨大提升空间；证据受限评估能同时提高准确率与可信度。
原文摘要 · Abstract (English)
Multimodal misinformation increasingly leverages visual persuasion, where repurposed or manipulated images strengthen misleading text. We introduce RW-Post, a post-aligned text--image benchmark for real-world multimodal fact-checking with auditable annotations: each instance links the original social-media post with reasoning traces and explicitly linked evidence items derived from human fact-check articles via an LLM-assisted extraction-and-auditing pipeline. RW-Post supports controlled evaluation across closed-book, evidence-bounded, and open-web regimes, enabling systematic diagnosis of visual grounding and evidence utilization. We provide AgentFact as a reference verification baseline and benchmark strong open-source LVLMs under unified protocols. Experiments show substantial headroom: current models struggle with faithful evidence grounding, while evidence-bounded evaluation improves both accuracy and faithfulness.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。