PulseAugur
中
实时 15:53:03
English(EN) ET-Prune: Evidence-Aware Dynamic Budgeting for Visual Token Pruning in Text-Rich MLLMs

ET-Prune 框架通过动态剪枝视觉令牌优化多模态大语言模型推理

研究人员开发了 ET-Prune,一个新颖的框架,旨在通过动态剪枝视觉令牌来优化多模态大语言模型(MLLMs)的推理成本。与固定的剪枝比例不同,ET-Prune 通过将剪枝视为证据分配来适应富文本输入,识别并保护查询指定的数字或标签等关键信息。该方法将证据不确定性和密度转化为特定于样本的令牌预算,从而在质量-成本权衡方面取得显著改进。在 OCRBench-v2 和 MMBench v1.1 上的评估中,ET-Prune 在剪枝方法中取得了具有竞争力或领先的结果,同时保留了大约一半的视觉令牌。 AI

影响 该方法可以显著降低运行多模态人工智能模型的计算成本,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多模态大语言模型新方法的论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CL 阅读 →

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

ET-Prune 框架通过动态剪枝视觉令牌优化多模态大语言模型推理

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
Tool
该集群包含一篇详细介绍多模态大语言模型新方法的论文。[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
59 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

报道来源 [1]

  1. arXiv cs.CL TIER_1 English(EN) · Zizhong Ding, Junxian Li, Kai Liu, Shaoqiu Zhang, Xiao Xiao, Linghe Kong, Yulun Zhang ·

    ET-Prune:面向文本密集型MLLM的证据感知动态预算视觉令牌剪枝

    arXiv:2608.01979v1 Announce Type: cross Abstract: Visual token pruning reduces the inference cost of multimodal large language models, but a fixed token ratio is poorly matched to text-rich inputs. In OCR-centric tasks, decisive evidence can be a small number, label, or field who…