HARK: Hierarchical Agentic Retrieval with Keyframing for Video Understanding (Student Abstract)
Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (48) · 41266-41268
DOI: 10.1609/aaai.v40i48.42237 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Jingcheng Li und 2 weiteren Autor:innen.
Abstract (Originalsprache)
Current video understanding models struggle with temporal reasoning and efficient processing while balancing detail preservation with computational efficiency. We propose a hierarchical memory system that segments videos into action and scene units, combined with question-aware agentic keyframe selection. Our method achieves 70.3% overall accuracy on VideoMME short video benchmarks.
Zitieren
APA
Li, J., Qiao, Y., & Huang, S. (2026). HARK: Hierarchical Agentic Retrieval with Keyframing for Video Understanding (Student Abstract). Proceedings of the AAAI Conference on Artificial Intelligence, 40(48), 41266-41268. https://doi.org/10.1609/aaai.v40i48.42237
Vancouver
Li J, Qiao Y, Huang S. HARK: Hierarchical Agentic Retrieval with Keyframing for Video Understanding (Student Abstract). Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(48):41266-41268. doi:10.1609/aaai.v40i48.42237
BibTeX
@article{Li2026,
author = {Li, Jingcheng and Qiao, Ye and Huang, Sitao},
title = {HARK: Hierarchical Agentic Retrieval with Keyframing for Video Understanding (Student Abstract)},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2026},
volume = {40},
number = {48},
pages = {41266--41268},
doi = {10.1609/aaai.v40i48.42237}
}
EUVIMED