English(EN) Sparse Text Diffusion under Dual Budgets: Context Connectivity and Expert Activation

新的DualBudget方法优化长上下文文本处理

作者 PulseAugur 编辑部 · [1 个来源] · 2026-11-02 05:00

研究人员引入了一种名为DualBudget的新方法来处理长上下文文本，该方法将计算成本分为两个独立的预算：注意力图的可用边和每个token激活的专家。该框架将各种注意力机制与条件专家计算相结合，旨在在不完全依赖稀疏性的情况下保留文档级信息。在摘要、问答、对话和句子对预测任务上的实验表明，在高达16K的上下文长度下，性能优于基线，并且与DiffuSeq相比，成本降低了20%。 AI

影响引入了一种高效长上下文文本处理的新方法，有望提高各种自然语言处理任务的性能并降低计算成本。

排序理由这是一篇详细介绍文本处理新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CL 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: 1 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 这是一篇详细介绍文本处理新方法的学术论文。[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: 1 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]

arXiv cs.CL TIER_1 English(EN) · Alexandros Christoforos, Chadbourne Davis, Paul Raj · 2026-11-02 05:00

Sparse Text Diffusion under Dual Budgets: Context Connectivity and Expert Activation

arXiv:2610.41487v1 Announce Type: new Abstract: Long-context diffusion repeatedly pays for both token interaction and feed-forward computation. We formulate these costs as two separate budgets: the edges available to an attention graph and the experts activated for each token. Th…

本文如何被排名

报道来源 [1]

Sparse Text Diffusion under Dual Budgets: Context Connectivity and Expert Activation

相关实体

相关话题