Insight-A: Attribution-aware for Multimodal Misinformation Detection

TL;DR

Insight-A introduces attribution-aware techniques leveraging hierarchical reasoning and prompting strategies to improve multimodal misinformation detection, addressing attribution challenges in AI-generated content.

Contribution

The paper proposes novel attribution-aware methods, including cross-attribution prompting and attribution-debiased prompting, for more accurate detection of multimodal misinformation.

Findings

Outperforms existing methods in misinformation detection accuracy

Effective attribution of forgery sources based on generation patterns

Enhanced cross-modal consistency checking with image captioning

Abstract

AI-generated content (AIGC) technology has emerged as a prevalent alternative to create multimodal misinformation on social media platforms, posing unprecedented threats to societal safety. However, standard prompting leverages multimodal large language models (MLLMs) to identify the emerging misinformation, which ignores the misinformation attribution. To this end, we present Insight-A, exploring attribution with MLLM insights for detecting multimodal misinformation. Insight-A makes two efforts: I) attribute misinformation to forgery sources, and II) an effective pipeline with hierarchical reasoning that detects distortions across modalities. Specifically, to attribute misinformation to forgery traces based on generation patterns, we devise cross-attribution prompting (CAP) to model the sophisticated correlations between perception and reasoning. Meanwhile, to reduce the subjectivity of human-annotated prompts, automatic attribution-debiased prompting (ADP) is used for task adaptation on MLLMs. Additionally, we design image captioning (IC) to achieve visual details for enhancing cross-modal consistency checking. Extensive experiments demonstrate the superiority of our proposal and provide a new paradigm for multimodal misinformation detection in the era of AIGC.