English(EN) NgramRL: Accelerating RL Post-Training for Structured Generation via Rollout-Aware N-gram Speculative Decoding

NgramRL通过感知回滚的N元语法推测解码加速LLM训练

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-26 04:00

研究人员开发了NgramRL，一种加速大型语言模型（LLM）强化学习（RL）训练后阶段的新颖方法。该技术解决了自回归回滚生成的高昂成本问题，该成本可能占训练时间的70%以上。NgramRL利用从近期回滚历史构建的每个提示的N元语法草稿生成器，无需辅助模型或持久GPU状态。自适应扇出预取器通过预测验证结果进一步优化性能，从而在每次前向传播的令牌承诺方面取得显著改进。 AI

影响 NgramRL可以显著降低训练大型语言模型的计算成本，可能使先进的RL训练后阶段更加易于获得。

排序理由该集群包含一篇详细介绍加速LLM训练新方法的 ist 研究论文。[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: 13 / 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加速LLM训练新方法的 ist 研究论文。[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: Same-day; Cluster formed today. Ranking reflects the current source set at time of score.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.AI TIER_1 English(EN) · Nilesh Prasad Pandey, Jangseon Park, Jason Kong, Onat Gungor, Tajana Rosing · 2026-10-26 04:00

NgramRL：通过感知滚动（Rollout-Aware）的N元语法（N-gram）推测解码加速结构化生成的强化学习（RL）后训练

arXiv:2610.32881v1 Announce Type: new Abstract: Reinforcement learning (RL) post-training improves the reasoning capabilities of large language models, but its cost is dominated by autoregressive rollout generation, which can exceed 70% of training time. Existing speculative-deco…

本文如何被排名

报道来源 [1]

NgramRL：通过感知滚动（Rollout-Aware）的N元语法（N-gram）推测解码加速结构化生成的强化学习（RL）后训练

相关实体

相关话题