English(EN) Prefilling-dLLM v2: Learned Chunk Routing for Long-Context Diffusion Language Models

Prefilling-dLLM v2 通过学习路由提升长上下文扩散模型性能

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-13 07:29

研究人员开发了Prefilling-dLLM v2，一种提高扩散语言模型在长上下文处理效率的方法。该方法涉及对模型进行持续训练，以便在块之间路由信息，而不仅仅是在推理时进行。通过引入学习路由槽和混合注意力机制，Prefilling-dLLM v2 在长达128K的RULER任务上显著优于其他稀疏方法，比次优稀疏模型准确率高出7.4个百分点。与全密集注意力相比，该优化还将解码速度提高了172倍。 AI

影响该方法显著增强了扩散模型的长上下文处理能力和解码速度，有望实现更高效的长文档或序列处理。

排序理由该集群描述了一篇关于改进扩散语言模型训练方法的新研究论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 Hugging Face Daily Papers 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; 该集群描述了一篇关于改进扩散语言模型训练方法的新研究论文。[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: 2 days old; Coverage has settled into its steady-state source set.

完整方法见我们的编辑标准。

报道来源 [1]

Hugging Face Daily Papers TIER_1 English(EN) · 2026-10-13 07:29

Prefilling-dLLM v2: 学习的块路由用于长上下文扩散语言模型

Diffusion language models apply bidirectional attention over the whole context at every denoising step, so a longer context makes each step substantially more expensive. Sparsifying only at inference leaves the pretrained weights unchanged, so the model is never trained under the…

本文如何被排名

报道来源 [1]

Prefilling-dLLM v2: 学习的块路由用于长上下文扩散语言模型

相关实体

相关话题