Metraipolis › archive › dc_d64193291a1f
Content-Aware Adaptive Token Pruning for Video-Language Models — arXiv:2608.03112
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.03112
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-31T12:18:33.951Z
- SHA-256 of the archived text
b8f849fc4a6b9d844af8d2c67762789bf74418e312510f398ce19ceac561ae7e- Stable URL of this page
https://metraipolis.com/documento/dc_d64193291a1f
How to cite
Content-Aware Adaptive Token Pruning for Video-Language Models — arXiv:2608.03112 — arXiv (en), https://arxiv.org/abs/2608.03112 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d64193291a1f (Metraipolis archive dc_d64193291a1f, sha256 b8f849fc4a6b9d844af8d2c67762789bf74418e312510f398ce19ceac561ae7e)
BibTeX
@misc{dcd64193291a1f,
title = {Content-Aware Adaptive Token Pruning for Video-Language Models — arXiv:2608.03112},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.03112},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d64193291a1f; sha256 b8f849fc4a6b9d844af8d2c67762789bf74418e312510f398ce19ceac561ae7e},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.03112. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Vision-language models excel at image and video understanding but suffer from high inference latency due to the need to process thousands of tokens per image, limiting their deployment on resource-constrained edge devices and in real-time surveillance applications. This challenge is further amplified in video processing, where multiple frames must be analyzed simultaneously. Existing token reduction techniques generally rely on a fixed, uniform pruning ratio applied across all inputs, which is suboptimal because the degree of redundancy can vary significantly between different videos, necessitating content-dependent pruning levels to preserve critical information. To address these limitations, we propose an adaptive token pruning strategy specifically designed for video processing. Crucially, the pruning ratio is determined adaptively based on the content of each video. This is achieved by analyzing the correlation structure of token embeddings to quantify redundancy, enabling dynamic token allocation that retains more tokens for dynamic scenes while aggressively pruning repetitive or static content. Importantly, our method is entirely post-hoc and requires no additional training or fine-tuning, while achieving strong empirical gains; notably, it improves accuracy by +7\% on a video captioning benchmark at 10\% token retention, while reducing computation TFLOPs by 95\%. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-05 in cs.AI). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2608.03112 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-31T12:05:52.998Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-05?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).