English(EN) Online Tokenization with Optimal Delay in Linear Time

新的在线分词算法大幅降低LLM延迟和内存使用

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-15 04:00

研究人员开发了一种新的在线分词算法，可显著降低大型语言模型（LLM）的延迟和内存使用量。该算法在近乎线性时间内实现了最优延迟，与OpenAI的tiktoken和Hugging Face的tokenizers等现有方法相比，具有显著的速度优势。实际应用证明，延迟提高了高达34.8倍，吞吐量提高了高达7倍，同时内存使用量保持不变。 AI

影响显著降低LLM的延迟和内存占用，可能实现更快、更高效的推理。

排序理由关于LLM分词新算法的学术论文。[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: 15 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 关于LLM分词新算法的学术论文。[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.LG TIER_1 English(EN) · Mina Dalirrooyfard, Christopher Ye · 2026-10-15 04:00

线性时间下的最优延迟在线分词

arXiv:2610.19139v1 Announce Type: cross Abstract: Tokenization is an essential preprocessing step in almost all large language models (LLMs). One of the most popular frameworks is BPE tokenization, which consists of special token segmentation, pretokenization, and byte-pair encod…

本文如何被排名

报道来源 [1]

线性时间下的最优延迟在线分词

相关实体

相关话题