arXiv preprint / cs.CV

Prune Once: Retraining-Free Task-Agnostic Pruning for Vision-Language Models

作者：Minseok Kang, Hyunwoo Kim, Chanyoung Kim, Minwoo Kim, Jaekoo Lee, Dahuin Jung · 会议/期刊：arXiv preprint · 方向：cs.CV · 发布日期：2026-08-10

2026-08-10视觉语言视觉感知
新标签打开 AlphaXiv新标签打开 Code
Prune Once: Retraining-Free Task-Agnostic Pruning for Vision-Language Models figure
AlphaXiv 中文概览（可滚动查看）