English(EN) How Model Growth, Recursion, and Boundary Operators Influence Scaling Exponents

Transformer架构调整提高了缩放效率，优于GPT-3

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-17 04:00

研究人员已经证明，对Transformer进行架构修改可以显著改变缩放指数，从而在相对于计算量的性能上实现指数级提升。通过引入递归深度等概念，例如“循环Transformer”和“边界算子”，模型可以实现更高的效率。一个利用模型增长的7.4B参数架构，以20倍的计算量优势匹配了13B参数GPT-3模型的性能，显示出效率增益随规模的增大而增大。 AI

影响新颖的架构技术可能导致更具计算效率的大型语言模型。

排序理由该条目是一篇学术论文，详细介绍了Transformer架构和缩放定律方面的新研究发现。[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: 12 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该条目是一篇学术论文，详细介绍了Transformer架构和缩放定律方面的新研究发现。[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.LG TIER_1 English(EN) · Zixi Chen, Akshay Vegesna, Samip Dahal, Andrew Gordon Wilson · 2026-09-17 04:00

模型增长、递归和边界算子如何影响缩放指数

arXiv:2609.19107v1 Announce Type: new Abstract: Scaling laws predict how loss decreases with increases in computation. We show, contrary to conventional wisdom, that architectural interventions can modify scaling exponents in pre-training, leading to exponential improvements in p…

本文如何被排名

报道来源 [1]

模型增长、递归和边界算子如何影响缩放指数

相关实体

相关话题