Omni-Fake: Benchmarking Unified Multimodal Social Media Deepfake Detection

TL;DR

Omni-Fake introduces a comprehensive multimodal deepfake dataset and a novel detection method, significantly improving robustness, generalization, and explainability in social media deepfake detection.

Contribution

The paper presents Omni-Fake, a large-scale multimodal dataset and a reinforcement-learning-based detector, advancing the evaluation and detection of deepfakes across multiple modalities.

Findings

Omni-Fake dataset contains over 1 million samples across four modalities.

The proposed Omni-Fake-R1 detector outperforms existing methods in accuracy and explainability.

Cross-modal generalization is significantly improved with the new dataset and detection framework.

Abstract

Multimodal deepfakes are proliferating on social media and threaten authenticity, information integrity, and digital forensics. Existing benchmarks are constrained by their single-modality scope, simplified manipulations, or unrealistic distributions, which limit their ability to assess real-world robustness. To address these limitations, we present Omni-Fake, a unified omni-dataset for comprehensive multimodal deepfake detection in social-media settings. It comprises Omni-Fake-Set, a large-scale, high-quality dataset with 1M+ samples, and Omni-Fake-OOD, an out-of-distribution benchmark with 200k+ samples intentionally excluded from training to evaluate generalization. Omni-Fake spans four modalities (image, audio, video, and audio-video talking head) and supports a joint detection-localization-explanation protocol. On top of Omni-Fake, we further propose Omni-Fake-R1, a reinforcement-learning-driven multimodal detector that adaptively integrates visual and auditory cues and outputs structured decisions, localization, and natural-language explanations. Extensive experiments show significant gains in detection accuracy, cross-modal generalization, and explainability over state-of-the-art baselines. Project page: https://tianxiao1201.github.io/omni-fake-project-page/