# GenVideoLens: Where LVLMs Fall Short in AI-Generated Video Detection?

arXiv: https://arxiv.org/abs/2603.18625 · HTML: https://mlsanity.com/paper/2603.18625/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 25 of its 36 references that are in the map.
- First posted on arXiv 19 Mar 2026.

## Builds on (most cited first, 25 of 25)

- [Open-Sora: Democratizing Efficient Video Production for All](https://mlsanity.com/paper/2412.20404/) (2024)
- [Sora: A Review on Background, Technology, Limitations, and Opportunities of Large Vision Models](https://mlsanity.com/paper/2402.17177/) (2024)
- [OpenVid-1M: A Large-Scale High-Quality Dataset for Text-to-video Generation](https://mlsanity.com/paper/2407.02371/) (2024)
- [DeMamba: AI-Generated Video Detection on Million-Scale GenVideo Benchmark](https://mlsanity.com/paper/2405.19707/) (2024)
- [Chain-of-Visual-Thought: Teaching VLMs to See and Think Better with Continuous Visual Tokens](https://mlsanity.com/paper/2511.19418/) (2025)
- [MMFakeBench: A Mixed-Source Multimodal Misinformation Detection Benchmark for LVLMs](https://mlsanity.com/paper/2406.08772/) (2024)
- [LOKI: A Comprehensive Synthetic Data Detection Benchmark using Large Multimodal Models](https://mlsanity.com/paper/2410.09732/) (2024)
- [On Learning Multi-Modal Forgery Representation for Diffusion Generated Video Detection](https://mlsanity.com/paper/2410.23623/) (2024)
- [FKA-Owl: Advancing Multimodal Fake News Detection through Knowledge-Augmented LVLMs](https://mlsanity.com/paper/2403.01988/) (2024)
- [FakeBench: Probing Explainable Fake Image Detection via Large Multimodal Models](https://mlsanity.com/paper/2404.13306/) (2024)
- [Ivy-Fake: A Unified Explainable Framework and Benchmark for Image and Video AIGC Detection](https://mlsanity.com/paper/2506.00979/) (2025)
- [D3: Training-Free AI-Generated Video Detection Using Second-Order Features](https://mlsanity.com/paper/2508.00701/) (2025)
- [BusterX: MLLM-Powered AI-Generated Video Forgery Detection and Explanation](https://mlsanity.com/paper/2505.12620/) (2025)
- [BusterX++: Towards Unified Cross-Modal AI-Generated Content Detection and Explanation with MLLM](https://mlsanity.com/paper/2507.14632/) (2025)
- [Video-SafetyBench: A Benchmark for Safety Evaluation of Video LVLMs](https://mlsanity.com/paper/2505.11842/) (2025)
- [Seeing What Matters: Generalizable AI-generated Video Detection with Forensic-Oriented Augmentation](https://mlsanity.com/paper/2506.16802/) (2025)
- [Beyond Deepfake Images: Detecting AI-Generated Videos](https://mlsanity.com/paper/2404.15955/) (2024)
- [Turns Out I'm Not Real: Towards Robust Detection of AI-Generated Videos](https://mlsanity.com/paper/2406.09601/) (2024)
- [Generative AI for Film Creation: A Survey of Recent Advances](https://mlsanity.com/paper/2504.08296/) (2025)
- [Forensics-Bench: A Comprehensive Forgery Detection Benchmark Suite for Large Vision Language Models](https://mlsanity.com/paper/2503.15024/) (2025)
- [DAVID-XR1: Detecting AI-Generated Videos with Explainable Reasoning](https://mlsanity.com/paper/2506.14827/) (2025)
- [REVEAL: Robust Evolution of Vision-Language Models for Explainable AI-Video Detection](https://mlsanity.com/paper/2502.14994/) (2025)
- [AEGIS: Authenticity Evaluation Benchmark for AI-Generated Video Sequences](https://mlsanity.com/paper/2508.10771/) (2025)
- [BrokenVideos: A Benchmark Dataset for Fine-Grained Artifact Localization in AI-Generated Videos](https://mlsanity.com/paper/2506.20103/) (2025)
- [AVFakeBench: A Comprehensive Audio-Video Forgery Detection Benchmark for AV-LMMs](https://mlsanity.com/paper/2511.21251/) (2025)

## Built on by (most cited first, 1 of 1)

- [Detect Early, Escalate Rarely: Anytime Detection of AI-Generated Video from the Compressed Bitstream](https://mlsanity.com/paper/2607.19476/) (2026)

## Abstract (arXiv)

In recent years, AI-generated videos have become increasingly realistic and sophisticated. Meanwhile, Large Vision-Language Models (LVLMs) have shown strong potential for detecting such content. However, existing evaluation protocols largely treat the task as a binary classification problem and rely on coarse-grained metrics such as overall accuracy, providing limited insight into where LVLMs succeed or fail. To address this limitation, we introduce GenVideoLens, a fine-grained benchmark that enables dimension-wise evaluation of LVLM capabilities in AI-generated video detection. The benchmark contains 400 highly deceptive AI-generated videos and 100 real videos, annotated by experts across 15 authenticity dimensions covering perceptual, optical, physical, and temporal cues. We evaluate eleven representative LVLMs on this benchmark. Our analysis reveals a pronounced dimensional imbalance. While LVLMs perform relatively well on perceptual cues, they struggle with optical consistency, physical interactions, and temporal-causal reasoning. Model performance also varies substantially across dimensions, with smaller open-source models sometimes outperforming stronger proprietary models on specific authenticity cues. Temporal perturbation experiments further show that current LVLMs make limited use of temporal information. Overall, GenVideoLens provides diagnostic insights into LVLM behavior, revealing key capability gaps and offering guidance for improving future AI-generated video detection systems.