English(EN) LearnPrunerV2: Learnable Progressive Token Pruning for Vision-Language Models

LearnPrunerV2通过可学习Token剪枝提升视觉语言模型效率

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-28 04:00

研究人员开发了LearnPrunerV2，一种用于视觉语言模型（VLMs）中渐进式剪枝Token的新方法，以降低计算开销。这种新方法将可学习模块集成到VLMs的每一层，通过结合视觉显著性和查询相关性来评估Token的重要性。LearnPrunerV2还引入了一种预算确定性剪枝策略，用于实现每层硬件友好的固定预算。实验表明，LearnPrunerV2在LLaVA-v1.5上显著优于现有方法，在剪枝77.8%的视觉Token的同时，达到了原始性能的99.8%。 AI

影响通过实现显著的Token减少而没有实质性的性能损失，提高了视觉语言模型的效率。

排序理由该集群描述了在arXiv上发布的一种用于改进视觉语言模型的新方法，属于研究范畴。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.AI 阅读 →

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: 15 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群描述了在arXiv上发布的一种用于改进视觉语言模型的新方法，属于研究范畴。[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: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.AI TIER_1 English(EN) · Rinyoichi Takezoe, Yaqian Li, Weiren Wang, Zihao Bo, Huihong Shi, Anzhou Hou, Mo Guang, Jiale Yan, Kaiwen Long · 2026-10-28 04:00

LearnPrunerV2：面向视觉语言模型的学习式渐进式Token剪枝

arXiv:2610.36986v1 Announce Type: cross Abstract: Vision-Language Models (VLMs) have demonstrated remarkable multimodal capabilities but suffer from prohibitive computational overhead due to the massive number of visual tokens. While token pruning is a promising solution, existin…

本文如何被排名

报道来源 [1]

LearnPrunerV2：面向视觉语言模型的学习式渐进式Token剪枝

相关实体

相关话题