Skyra: AI-Generated Video Detection via Grounded Artifact Reasoning

TL;DR

Skyra is a multimodal large language model designed to detect AI-generated videos by identifying visual artifacts and providing human-understandable explanations, supported by a new large-scale dataset and benchmark.

Contribution

The paper introduces Skyra, a novel model that detects and explains AI-generated videos using grounded artifact reasoning, along with a large dataset and benchmark for evaluation.

Findings

Skyra outperforms existing detection methods on multiple benchmarks.

The model provides human-interpretable explanations for its detections.

The new dataset ViF-CoT-4K enables detailed artifact annotation for training.

Abstract

The misuse of AI-driven video generation technologies has raised serious social concerns, highlighting the urgent need for reliable AI-generated video detectors. However, most existing methods are limited to binary classification and lack the necessary explanations for human interpretation. In this paper, we present Skyra, a specialized multimodal large language model (MLLM) that identifies human-perceivable visual artifacts in AI-generated videos and leverages them as grounded evidence for both detection and explanation. To support this objective, we construct ViF-CoT-4K for Supervised Fine-Tuning (SFT), which represents the first large-scale AI-generated video artifact dataset with fine-grained human annotations. We then develop a two-stage training strategy that systematically enhances our model's spatio-temporal artifact perception, explanation capability, and detection accuracy. To comprehensively evaluate Skyra, we introduce ViF-Bench, a benchmark comprising 3K high-quality samples generated by over ten state-of-the-art video generators. Extensive experiments demonstrate that Skyra surpasses existing methods across multiple benchmarks, while our evaluation yields valuable insights for advancing explainable AI-generated video detection.