ALLVB: All-in-One Long Video Understanding Benchmark
Proceedings of the AAAI Conference on Artificial Intelligence · 11.04.2025 · 39 (7) · 7211-7219
DOI: 10.1609/aaai.v39i7.32775 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 11.04.2025. Verfasst von Xichen Tan und 4 weiteren Autor:innen. Die Arbeit wurde bisher 7-mal zitiert (Crossref).
Abstract (Originalsprache)
From image to video understanding, the capabilities of Multi-modal LLMs (MLLMs) are increasingly powerful. However, most existing video understanding benchmarks are relatively short, which makes them inadequate for effectively evaluating the long-sequence modeling capabilities of MLLMs. This highlights the urgent need for a comprehensive and integrated long video understanding benchmark to assess the ability of MLLMs thoroughly. To this end, we propose ALLVB (ALL-in-One Long Video Understanding Benchmark). ALLVB's main contributions include: 1) It integrates 9 major video understanding tasks. These tasks are converted into video QA formats, allowing a single benchmark to evaluate 9 different video understanding capabilities of MLLMs, highlighting the versatility, comprehensiveness, and challenging nature of ALLVB. 2) A fully automated annotation pipeline using GPT-4o is designed, requiring only human quality control, which facilitates the maintenance and expansion of the benchmark. 3) It contains 1,376 videos across 16 categories, averaging nearly 2 hours each, with a total of 252k QAs. To the best of our knowledge, it is the largest long video understanding benchmark in terms of the number of videos, average duration, and number of QAs. We have tested various mainstream MLLMs on ALLVB, and the results indicate that even the most advanced commercial models have significant room for improvement. This reflects the benchmark's challenging nature and demonstrates the substantial potential for development in long video understanding.
Zitieren
APA
Tan, X., Luo, Y., Ye, Y., Liu, F., & Cai, Z. (2025). ALLVB: All-in-One Long Video Understanding Benchmark. Proceedings of the AAAI Conference on Artificial Intelligence, 39(7), 7211-7219. https://doi.org/10.1609/aaai.v39i7.32775
Vancouver
Tan X, Luo Y, Ye Y, Liu F, Cai Z. ALLVB: All-in-One Long Video Understanding Benchmark. Proceedings of the AAAI Conference on Artificial Intelligence. 2025;39(7):7211-7219. doi:10.1609/aaai.v39i7.32775
BibTeX
@article{Tan2025,
author = {Tan, Xichen and Luo, Yuanjing and Ye, Yunfan and Liu, Fang and Cai, Zhiping},
title = {ALLVB: All-in-One Long Video Understanding Benchmark},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2025},
volume = {39},
number = {7},
pages = {7211--7219},
doi = {10.1609/aaai.v39i7.32775}
}
EUVIMED