PulseAugur
中
实时 21:40:27
English(EN) MergeOver: Post-Training Token Merging for Recursive Vision Transformers

新的方法出现，用于AI模型中高效的视觉Token修剪 · 跟踪6个来源

作者 PulseAugur 编辑部 · [8 个来源] ·

研究人员正在开发新的方法来优化视觉Transformer (ViTs) 和多模态大语言模型 (MLLMs)，通过修剪计算成本高昂的视觉Token。几篇论文提出了新颖的技术，以在不显著降低性能的情况下减少处理的Token数量。这些方法包括递归ViTs的训练后Token合并、使用专用寄存器的任务自适应修剪、预测中间层注意力以确定Token重要性、面向角色的区域分配以及用于设计修剪策略的AI驱动框架。 AI

影响 这些进展旨在显著降低大型AI模型的计算成本和延迟，从而能够在资源受限的设备上更广泛地部署。

排序理由 多篇发表在arXiv上的研究论文详细介绍了AI模型中视觉Token修剪的新颖方法。

在 arXiv cs.LG 阅读 →

AI 生成摘要 · Google Gemini · 来自 8 个来源。 我们如何撰写摘要 →

新的方法出现，用于AI模型中高效的视觉Token修剪 · 跟踪6个来源

本文如何被排名

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Research
多篇发表在arXiv上的研究论文详细介绍了AI模型中视觉Token修剪的新颖方法。
Source corroboration
8 independent sources
Strong cross-source corroboration — multiple independent publishers covered this within the clustering window.
Topics
model release, 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
50 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.
Coverage growth since scoring
+2 source(s) since last score
New sources have picked up this story since our last re-score. Score will update on the next scoring pass.

完整方法见我们的编辑标准。

报道来源 [8]

  1. arXiv cs.LG TIER_1 English(EN) · Junseo Kim, Uraz Odyurt, Amirreza Yousefzadeh ·

    MergeOver：递归视觉Transformer的训练后Token合并

    arXiv:2608.13141v1 Announce Type: cross Abstract: Vision Transformers (ViTs) demonstrate exceptional performance in computer vision but suffer from large parameter counts and quadratic computational complexity, severely limiting their deployment on resource-constrained edge hardw…

  2. arXiv cs.AI TIER_1 Dansk(DA) · Hongsen Cao, Mona Jaber, Shanxin Yuan, Ahmed Sayed ·

    让寄存器发挥作用：用于 Vision Transformer 中 Token 修剪的任务寄存器

    arXiv:2608.10989v1 Announce Type: cross Abstract: Token-pruning policies are usually designed for a single recognition pipeline, but pretrained Vision Transformers are reused across tasks with different spatial demands. We ask which parts of a pruning policy transfer across image…

  3. Hugging Face Daily Papers TIER_1 English(EN) ·

    当视觉化为文本：通过跨模态残差引导在视觉语言模型中进行视觉标记剪枝

    Abundant visual information strengthens vision-language model (VLM) perception, yet massive visual tokens raise inference costs. Existing visual token pruning methods rely on similarity-based guidance, which exploits pairwise text-vision and vision-vision token correlations for c…

  4. arXiv cs.AI TIER_1 English(EN) · Yuyao Sun, Tao Deng, Shuang Li, Deqing Wang, Hao Geng, Minjun Yu ·

    为视觉Token修剪学习预测MLLMs中的中间层注意力

    arXiv:2608.06411v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) achieve strong performance across diverse vision-language tasks, but their efficiency is limited by the cost of processing numerous visual tokens. Visual token pruning can reduce this cost, b…

  5. arXiv cs.AI TIER_1 English(EN) · Qiyanhui Lu, Han Wu, Rongjian Xu, Tingzhang Luo, Cheng Fan, Xinghao Chen, Minjing Dong, Jufeng Yang, Jianyuan Guo ·

    RoRA：面向多模态大语言模型视觉令牌剪枝的角色导向区域分配

    arXiv:2608.07088v1 Announce Type: cross Abstract: Multimodal large language models (MLLMs) encode images as long visual token sequences, making prefilling and KV-cache storage expensive. Existing training-free pruning methods select tokens by importance, diversity, or spatial cov…

  6. arXiv cs.LG TIER_1 English(EN) · Zhen Liu, Wenli Huang, Wei Song, Yuhan Liu, Zhiqin Yang, Jingwen Fu ·

    面向视觉令牌剪枝的AI4AI框架

    arXiv:2608.07193v1 Announce Type: new Abstract: Visual-token pruning can substantially reduce the inference cost of multimodal large language models (MLLMs), yet existing methods largely rely on fixed, handcrafted heuristics and costly expert trial and error. As pruning objective…

  7. Hugging Face Daily Papers TIER_1 English(EN) ·

    面向视觉令牌剪枝的AI4AI框架

    AutoPrune uses large language models to automatically design visual-token pruning policies for multimodal models via a domain-specific language and residual search formulation, achieving high efficiency with minimal performance loss.

  8. arXiv cs.CV TIER_1 English(EN) · Congyang Ou, Ruike Song, Yang Zhou, Libo Sun, Haokui Zhang, Zhenbo Luo ·

    当视觉变为文本：通过跨模态残差引导在视觉语言模型中进行视觉标记剪枝

    arXiv:2608.10489v1 Announce Type: new Abstract: Abundant visual information strengthens vision-language model (VLM) perception, yet massive visual tokens raise inference costs. Existing visual token pruning methods rely on similarity-based guidance, which exploits pairwise text-v…

相关实体

相关话题