Paper
GenVideoLens: Where LVLMs Fall Short in AI-Generated Video Detection?
Authors
Yueying Zou, Pei Pei Li, Zekun Li, Xinyu Guo, Xing Cui, Huaibo Huang, Ran He
Abstract
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.
Metadata
Related papers
Vibe Coding XR: Accelerating AI + XR Prototyping with XR Blocks and Gemini
Ruofei Du, Benjamin Hersh, David Li, Nels Numan, Xun Qian, Yanhe Chen, Zhongy... • 2026-03-25
Comparing Developer and LLM Biases in Code Evaluation
Aditya Mittal, Ryan Shar, Zichu Wu, Shyam Agarwal, Tongshuang Wu, Chris Donah... • 2026-03-25
The Stochastic Gap: A Markovian Framework for Pre-Deployment Reliability and Oversight-Cost Auditing in Agentic Artificial Intelligence
Biplab Pal, Santanu Bhattacharya • 2026-03-25
Retrieval Improvements Do Not Guarantee Better Answers: A Study of RAG for AI Policy QA
Saahil Mathur, Ryan David Rittner, Vedant Ajit Thakur, Daniel Stuart Schiff, ... • 2026-03-25
MARCH: Multi-Agent Reinforced Self-Check for LLM Hallucination
Zhuo Li, Yupeng Zhang, Pengyu Cheng, Jiajun Song, Mengyu Zhou, Hao Li, Shujie... • 2026-03-25
Raw Data (Debug)
{
"raw_xml": "<entry>\n <id>http://arxiv.org/abs/2603.18625v1</id>\n <title>GenVideoLens: Where LVLMs Fall Short in AI-Generated Video Detection?</title>\n <updated>2026-03-19T08:44:08Z</updated>\n <link href='https://arxiv.org/abs/2603.18625v1' rel='alternate' type='text/html'/>\n <link href='https://arxiv.org/pdf/2603.18625v1' rel='related' title='pdf' type='application/pdf'/>\n <summary>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.</summary>\n <category scheme='http://arxiv.org/schemas/atom' term='cs.CV'/>\n <published>2026-03-19T08:44:08Z</published>\n <arxiv:comment>ECCV 2026 submission. 14 pages, 6 figures, 4 tables. Supplementary material included</arxiv:comment>\n <arxiv:primary_category term='cs.CV'/>\n <author>\n <name>Yueying Zou</name>\n </author>\n <author>\n <name>Pei Pei Li</name>\n </author>\n <author>\n <name>Zekun Li</name>\n </author>\n <author>\n <name>Xinyu Guo</name>\n </author>\n <author>\n <name>Xing Cui</name>\n </author>\n <author>\n <name>Huaibo Huang</name>\n </author>\n <author>\n <name>Ran He</name>\n </author>\n </entry>"
}