LAVID: An Agentic LVLM Framework for Diffusion-Generated Video Detection

TL;DR

LAVID introduces a training-free, LVLM-based framework for detecting AI-generated videos by leveraging external tools and adaptive prompt structuring, significantly improving detection accuracy on a new benchmark.

Contribution

The paper presents LAVID, a novel LVLM-based, training-free video detection method that uses explicit knowledge and adaptive prompts, addressing limitations of traditional deep learning approaches.

Findings

LAVID improves F1 scores by 6.2 to 30.2% over top baselines.

The method is fully inference-based, avoiding additional training.

Evaluation on the new VidFor dataset demonstrates effectiveness.

Abstract

The impressive achievements of generative models in creating high-quality videos have raised concerns about digital integrity and privacy vulnerabilities. Recent works of AI-generated content detection have been widely studied in the image field (e.g., deepfake), yet the video field has been unexplored. Large Vision Language Model (LVLM) has become an emerging tool for AI-generated content detection for its strong reasoning and multimodal capabilities. It breaks the limitations of traditional deep learning based methods faced with like lack of transparency and inability to recognize new artifacts. Motivated by this, we propose LAVID, a novel LVLMs-based ai-generated video detection with explicit knowledge enhancement. Our insight list as follows: (1) The leading LVLMs can call external tools to extract useful information to facilitate its own video detection task; (2) Structuring the prompt can affect LVLM's reasoning ability to interpret information in video content. Our proposed pipeline automatically selects a set of explicit knowledge tools for detection, and then adaptively adjusts the structure prompt by self-rewriting. Different from prior SOTA that trains additional detectors, our method is fully training-free and only requires inference of the LVLM for detection. To facilitate our research, we also create a new benchmark \vidfor with high-quality videos generated from multiple sources of video generation tools. Evaluation results show that LAVID improves F1 scores by 6.2 to 30.2% over the top baselines on our datasets across four SOTA LVLMs.