State-Space Hierarchical Compression with Gated Attention and Learnable Sampling for Hour-Long Video Understanding in Large Multimodal Models
Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (7) · 5656-5664
DOI: 10.1609/aaai.v40i7.37485 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Geewook Kim und 1 weiteren Autor:innen.
Abstract (Originalsprache)
We propose an efficient framework to compress massive video-frame features before feeding them into large multimodal models, thereby mitigating the severe token explosion arising from hour-long videos. Our design leverages a bidirectional state-space model equipped with a gated skip connection and a learnable weighted-average pooling mechanism applied to periodically inserted learned queries. This structure enables hierarchical downsampling across both spatial and temporal dimensions, preserving performance in a cost-effective manner. Across challenging hour-long video understanding tasks, our approach demonstrates competitive results against state-of-the-art models, while significantly reducing overall token budget. Notably, replacing our state-space model with conventional modules results in substantial performance degradation, highlighting the advantages of the proposed state-space modeling for effectively compressing multi-frame video information. Our framework emphasizes resource-conscious efficiency, making it practical for real-world deployments. We validate its scalability and generality across multiple benchmarks, achieving the dual objectives of efficient resource usage and comprehensive video understanding.
Zitieren
APA
Kim, G., & Seo, M. (2026). State-Space Hierarchical Compression with Gated Attention and Learnable Sampling for Hour-Long Video Understanding in Large Multimodal Models. Proceedings of the AAAI Conference on Artificial Intelligence, 40(7), 5656-5664. https://doi.org/10.1609/aaai.v40i7.37485
Vancouver
Kim G, Seo M. State-Space Hierarchical Compression with Gated Attention and Learnable Sampling for Hour-Long Video Understanding in Large Multimodal Models. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(7):5656-5664. doi:10.1609/aaai.v40i7.37485
BibTeX
@article{Kim2026,
author = {Kim, Geewook and Seo, Minjoon},
title = {State-Space Hierarchical Compression with Gated Attention and Learnable Sampling for Hour-Long Video Understanding in Large Multimodal Models},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2026},
volume = {40},
number = {7},
pages = {5656--5664},
doi = {10.1609/aaai.v40i7.37485}
}
EUVIMED