Misinformation is a prevalent societal issue due to its potential high risks.
Out-of-context (OOC) misinformation, where authentic images are repurposed with
false text, is one of the easiest and most effective ways to mislead audiences.
Current methods focus on assessing image-text consistency but lack convincing
explanations for their judgments, which is essential for debunking
misinformation. While Multimodal Large Language Models (MLLMs) have rich
knowledge and innate capability for visual reasoning and explanation
generation, they still lack sophistication in understanding and discovering the
subtle crossmodal differences. In this paper, we introduce SNIFFER, a novel
multimodal large language model specifically engineered for OOC misinformation
detection and explanation. SNIFFER employs two-stage instruction tuning on
InstructBLIP. The first stage refines the model's concept alignment of generic
objects with news-domain entities and the second stage leverages language-only
GPT-4 generated OOC-specific instruction data to fine-tune the model's
discriminatory powers. Enhanced by external tools and retrieval, SNIFFER not
only detects inconsistencies between text and image but also utilizes external
knowledge for contextual verification. Our experiments show that SNIFFER
surpasses the original MLLM by over 40% and outperforms state-of-the-art
methods in detection accuracy. SNIFFER also provides accurate and persuasive
explanations as validated by quantitative and human evaluations.

SNIFFER 是一种特别针对超文本误导检测和解释的新型多模态大型语言模型，通过在 InstructBLIP 上进行两阶段的指令调整，不仅能够检测文字和图像之间的不一致性，还可以利用外部知识进行上下文验证，实验证明 SNIFFER 的检测准确性超过了原始的 MLLM 超过 40%，并且在解释能力方面也优于最先进的方法。