新的DASC方法将AI模型状态压缩率提高2.63倍

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-01 04:00

研究人员开发了衰减感知状态压缩（DASC）方法，这是一种优化混合线性注意力模型服务的新颖方法。DASC分析不同模型组件的保留时间尺度，识别哪些状态部分可以在不显著损失质量的情况下进行压缩。通过选择性地存储和打包长时程状态单元，DASC可以将循环状态检查点的内存使用量减少高达2.63倍。这种压缩带来了推理速度的显著提升，包括平均首次令牌时间（Time to First Token）减少42.6%，以及输入吞吐量增加68.4%。 AI

影响该技术显著提高了大型语言模型服务的效率，有可能降低基础设施成本并提高可访问性。

排序理由该集群包含一篇详细介绍AI模型服务新技术的学术论文。[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: 30 / 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AI模型服务新技术的学术论文。[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: 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) · Yanqi Yu, Pingwei Sun, Jianchao Tan, Tao Zhang, Yuchen Xie, Xunliang Cai, Yao Liu · 2026-09-01 04:00

DASC：混合线性注意力服务中的衰减感知状态压缩

arXiv:2608.30386v1 Announce Type: cross Abstract: Hybrid linear-attention architectures have recently scaled to large open-weight models, offering quality competitive with full attention while substantially reducing key/value (KV) cache growth. However, their in-place recurrent-s…

本文如何被排名

报道来源 [1]

DASC：混合线性注意力服务中的衰减感知状态压缩

相关实体

相关话题