Search papers, labs, and topics across Lattice.
This paper introduces SRM-FND, a self-reflective multimodal reasoning framework designed for detecting fake news in short videos, addressing key challenges in reasoning quality and model fine-tuning without relying on ground-truth supervision. The framework employs techniques such as contrastive deliberation and iterative root-cause diagnosis, alongside a collaborative approach involving a Blind Analyst and Counter-Conclusion Reasoner to enhance the identification of discriminative rationales. Experimental results on the FakeSV and FakeTT datasets demonstrate that SRM-FND significantly outperforms existing baselines, yielding more reliable predictions and improved performance across different datasets.
SRM-FND achieves superior fake news detection by enhancing reasoning quality through self-reflection, outperforming traditional methods in reliability and interpretability.
Recent fake news detection pipelines increasingly leverage large language models and vision-language models for reasoning-based analysis. However, several challenges remain open: improving reasoning quality through self-reflection without ground-truth chain-of-thought supervision, using improved reasoning to benefit downstream model fine-tuning, and connecting single-sample fraudulent-pattern discovery with cross-sample verification. We propose SRM-FND, a self-reflective multimodal reasoning framework for short-video fake news detection. SRM-FND develops higher-quality reasoning through contrastive deliberation, iterative root-cause diagnosis, and corrective prompt refinement. A Blind Analyst, Counter-Conclusion Reasoner, and Self-Consistency Arbiter collaboratively identify and retain discriminative rationales. The framework also incorporates dual-phase, topic-adaptive vision-language model fine-tuning to improve multimodal grounding and enable lightweight topic specialization. For uncertain cases, it performs confidence-driven cross-sample review by retrieving credible and suspicious co-event examples. Experiments on FakeSV and FakeTT show that SRM-FND outperforms strong baselines, produces more reliable and interpretable predictions, and delivers noticeable improvements in cross-dataset performance.
