BusterX: MLLM-Powered AI-Generated Video Forgery Detection and Explanation

TL;DR

The paper tackles the rising threat of AI-generated video forgery by introducing GenBuster-200K, a large-scale, fair, real-world video dataset, and BusterX, an MLLM-powered detector that provides explanations via reinforcement learning. BusterX reframes detection as visual reasoning with step-by-step CoT and trainable explanations, guided by a Dynamic Sampling Policy Optimization reward framework. The authors demonstrate state-of-the-art performance and strong out-of-domain generalization, supported by a Closed Benchmark to assess unseen generators. This work advances both practical detection capabilities and the interpretability of AI-generated video analysis, with significant implications for public trust and misinformation mitigation.

Abstract

Advances in AI generative models facilitate super-realistic video synthesis, amplifying misinformation risks via social media and eroding trust in digital content. Several research works have explored new deepfake detection methods on AI-generated images to alleviate these risks. However, with the fast development of video generation models, such as Sora and WanX, there is currently a lack of large-scale, high-quality AI-generated video datasets for forgery detection. In addition, existing detection approaches predominantly treat the task as binary classification, lacking explainability in model decision-making and failing to provide actionable insights or guidance for the public. To address these challenges, we propose \textbf{GenBuster-200K}, a large-scale AI-generated video dataset featuring 200K high-resolution video clips, diverse latest generative techniques, emphasis on fairness, and focus on real-world scenes. We further introduce \textbf{BusterX}, a novel AI-generated video detection and explanation framework leveraging multimodal large language model (MLLM) and reinforcement learning (RL) to provide authenticity determination and explainable rationales. To our knowledge, BusterX is the first framework to integrate MLLM with RL for explainable AI-generated video detection. Extensive experiments with state-of-the-art methods and ablation studies demonstrate the effectiveness and generalizability of BusterX.

Figures (13)

• Figure 1: Benchmark Construction of GenBuster-200K.
• Figure 2: BusterX utilizes MLLM reasoning to analyze a video and determine if it is AI-generated, the reasoning process itself serves as the detection mechanism. More details are in the appendix.
• Figure 3: Visual Examples from GenBuster-200K.
• Figure 4: Case Study. Full responses are provided in appendix.
• Figure 5: CoT Length comparison across different reward functions.