新的QViD方法修剪VLMs中的视觉标记以提高效率

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-27 04:00

研究人员推出了一种名为查询-视觉分解（QViD）的新方法，用于修剪大型视觉语言模型（VLMs）中的视觉标记。这种无需训练的方法利用查询-视觉交互矩阵的低秩结构来识别和移除不太相关的视觉标记，从而提高效率。在图像和视频理解基准上的实验表明，QViD能够提高性能，尤其是在激进的压缩设置下。 AI

影响该方法可以显著降低大型视觉语言模型的推理成本，使其在各种应用中更易于访问和更高效。

排序理由该集群包含一篇详细介绍提高AI模型效率的新方法的论文。[lever_c_demoted from research: ic=1 ai=1.0]

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 10 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇详细介绍提高AI模型效率的新方法的论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, infra; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Same-day; Cluster formed today. Ranking reflects the current source set at time of score.

完整方法见我们的编辑标准。

arXiv cs.CV TIER_1 English(EN) · Harshithanjani Athi, Sravan Kumar Ankireddy, Jianzhong Charlie Zhang, Hyeji Kim · 2026-10-27 04:00

通过查询-视觉交互分解实现视觉令牌剪枝

arXiv:2610.33215v1 Announce Type: cross Abstract: Large vision-language models (VLMs), which process both visual inputs and text queries, incur high inference costs due to the large number of visual tokens, making visual token pruning a natural approach for improving efficiency. …