PulseAugur
中
实时 20:09:49
English(EN) Why Do Accumulated Transformations Extrapolate?

累积变换可改善LLM长度外推能力,但在极端情况下会下降

研究人员调查了累积变换在注意力机制中的外推能力,特别研究了用累积的、依赖数据的Householder反射替换RoPE的位置索引旋转如何影响性能。他们的发现表明,虽然这些累积变换可以改善长度外推能力,但在极端上下文长度下性能最终会下降。该研究还探讨了一种使用累积的、依赖令牌的旋转的简化变体,该变体表现出类似的行为。理论分析表明,累积的正交变换在有限步数后会导致不连贯,限制了对远距离令牌的注意力,并创建了一个有限的混合窗口。 AI

影响 调查了当前注意力机制在处理极端上下文长度方面的局限性,可能指导未来的架构改进。

排序理由 学术论文,详细介绍了注意力机制的理论和实验结果。[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 →

累积变换可改善LLM长度外推能力,但在极端情况下会下降

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
学术论文,详细介绍了注意力机制的理论和实验结果。[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, model release
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
105 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]

  1. arXiv cs.CL TIER_1 English(EN) · Mahesh Godavarti ·

    累积的变换为何能外推?

    arXiv:2606.24975v1 Announce Type: cross Abstract: PaTH Attention showed that replacing RoPE's position-indexed rotations with accumulated data-dependent Householder reflections yields strong length extrapolation, though performance degrades at extreme context lengths. We ask whet…