PulseAugur
中
实时 09:13:56
English(EN) GSM: Efficient Language Modeling with Shared Global State

新的全局状态模型架构提高了语言模型的效率

研究人员推出了一种新颖的因果编码器-解码器架构——全局状态模型(GSM),旨在提高语言模型的效率。GSM 通过使用编码器将长程信息聚合到共享状态中,然后解码器层在不重复访问历史记录的情况下进行处理,从而将历史信息获取与深度计算分离开来。这种方法使得解码器的注意力(attention)和缓存成本与历史长度无关,而依赖感知剪枝(dependency-aware pruning)则减少了预填充(prefill)计算。实验表明,GSM 在达到与 DeepSeek-V4.1-Flash 基线相当的性能的同时,在某些规模下具有改进的潜力,并且还提高了训练和推理吞吐量并降低了缓存开销。 AI

影响 这种新架构可能导致更高效的大型语言模型的训练和推理,降低计算成本并提高长上下文任务的性能。

排序理由 该集群包含一篇详细介绍新模型架构的学术论文。[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 →

新的全局状态模型架构提高了语言模型的效率

本文如何被排名

Signal score
14 / 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新模型架构的学术论文。[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
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

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) · Yunao Zheng, Bin Wen, Xiaojie Wang, Kaiyu Jiang, Xuanyu Zheng, Changyi Liu, Hongyi Fu, Jianxiong Wang, Tianke Zhang, Haonan Fan, Yingxin Li, Jiankang Chen, Xu Wang, Tingting Gao, Han Li ·

    GSM:具有共享全局状态的高效语言建模

    arXiv:2609.33465v3 Announce Type: replace Abstract: Improving language model efficiency requires not only cheaper operations but also eliminating redundant computation. We argue that effective modeling requires information accessibility and computational depth, but not repeated h…