Metraipolis › archive › dc_cea554784093
BtsPruner: Role-Aware Visual Token Pruning for Large Vision-Language Models — arXiv:2611.04537
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.04537
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-04T12:29:57.152Z
- SHA-256 of the archived text
d79668d2ab111409e6824317e3d68dcc0c75dc61d2005f394dbb15a81afdac9e- Stable URL of this page
https://metraipolis.com/documento/dc_cea554784093
How to cite
BtsPruner: Role-Aware Visual Token Pruning for Large Vision-Language Models — arXiv:2611.04537 — arXiv (en), https://arxiv.org/abs/2611.04537 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cea554784093 (Metraipolis archive dc_cea554784093, sha256 d79668d2ab111409e6824317e3d68dcc0c75dc61d2005f394dbb15a81afdac9e)
BibTeX
@misc{dccea554784093,
title = {BtsPruner: Role-Aware Visual Token Pruning for Large Vision-Language Models — arXiv:2611.04537},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.04537},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cea554784093; sha256 d79668d2ab111409e6824317e3d68dcc0c75dc61d2005f394dbb15a81afdac9e},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.04537. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Large Vision-Language Models (LVLMs) often process hundreds or thousands of visual tokens, resulting in substantial computational and memory overhead during inference. Existing training-free pruning methods typically estimate token importance using language-model attention, visual attention, or feature similarity. However, these methods treat visual tokens with uniform criteria and overlook that tokens may serve different roles during visual encoding. In this work, we investigate the evolution of visual token roles in LVLMs and observe that tokens progressively transition from local detail carriers to globally referenced representatives as encoding proceeds. Based on this observation, we propose BtsPruner, a training-free role-aware visual token pruning framework that preserves complementary token roles under a fixed budget. Specifically, BtsPruner first identifies the token-role transition point based on the evolution of attention concentration. It then discovers globally referenced representatives from later encoding stages and recursively retrieves local detail tokens from earlier stages to recover complementary fine-grained evidence. Extensive experiments across LLaVA-1.5, LLaVA-NeXT, Qwen-VL, InternVL, and CogVLM demonstrate that BtsPruner achieves superior performance retention under aggressive token reduction. On LLaVA-1.5-7B, BtsPruner preserves 91.6% of the dense model's performance using only 32 of 576 visual tokens (5.6%). In the high-resolution LLaVA-NeXT-7B setting, it achieves up to a 5.95x CUDA-latency speedup. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2611.04537 — 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 2026-11-04T12:26:47.280Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).