Decoupling Consistency and Conflict across MLLM Layers for Fake News Detection

arXiv:2610.36569v1 Announce Type: cross Abstract: The trustworthiness of Web content faces a growing threat from multimodal fake news. Real news usually shows aligned image-text semantics, while fake news often contains a few conflicts at varying levels, such as mismatches in objects, scenes, events, and claims. However, existing methods usually treat consistency and conflict in a coarse and static way, limiting their ability to capture sparse conflicts and adapt to varying conflict levels. Inspired by the distinct knowledge encoded across layers in large language models, we propose Multi-Level Consistency and Conflict Decoupling (MLCCD), which captures conflicts at different levels from different layers of multimodal large language models (MLLMs). Specifically, we first use the MLLM's final decision representation to select visual and textual tokens relevant to the task, preventing irrelevant content from diluting sparse conflicts. Then, we preserve tokens whose representations change sharply between adjacent layers, capturing the knowledge each layer adds. Finally, we decouple this knowledge into consistency and conflict evidence and aggregate each across layers with its own router, adapting to varying conflict levels. Experiments on three benchmarks show that MLCCD outperforms existing methods.

Related

• RAEGNet: Relation-Aware Evidence Graph Network for Harm-Aware Multimodal Fake News Detection
• From Generation to Detection: Exploration of Discourse Driven Scenario based LLM Generated Fake News
• Same Semantics, Different Outcome: On the Modality Robustness of Multimodal LLMs under Knowledge Conflict

Source: arXiv cs.CL | 2026-10-28