English(EN) BlockServe: Block-Grained Continuous Batching for High-Throughput Diffusion LLM Serving

BlockServe框架将dLLM服务吞吐量提升高达10.6倍

作者 PulseAugur 编辑部 · [1 个来源] · 2026-07-13 04:00

研究人员开发了BlockServe，一个旨在提高扩散式大语言模型（dLLM）服务效率的新框架。该系统通过实现块粒度调度来解决批处理中的收敛异构性挑战，该调度允许在块边界立即驱逐已完成的请求。BlockServe还利用混合状态执行和计算感知准入控制器来扩大有效批容量。在各种基准测试中对Dream和LLaDA模型进行的实验表明，与Fast-dLLM相比，BlockServe的吞吐量可提高1.9倍至10.6倍，同时保持相似的生成质量。 AI

影响该框架可以显著提高大型语言模型的部署效率并降低延迟，特别是对于基于扩散的模型。

排序理由该集群包含一篇详细介绍LLM服务新技术的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.LG 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: 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LLM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: infra, paper; 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: 46 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]

arXiv cs.LG TIER_1 English(EN) · Yuanjie Zhu, Liangwei Yang, Ke Xu, Weizhi Zhang, Shanghao Li, Zihe Song, Philip S. Yu · 2026-07-13 04:00

BlockServe：面向高吞吐量扩散式大模型服务的块级连续批处理

arXiv:2607.08930v1 Announce Type: new Abstract: Efficient serving of diffusion large language models (dLLMs) is hindered by convergence heterogeneity: when batching multiple requests, different sequences converge at different rates, causing faster requests to stall behind slower …

本文如何被排名

报道来源 [1]

BlockServe：面向高吞吐量扩散式大模型服务的块级连续批处理

相关实体

相关话题