English(EN) Does the Linguistic Register of Thinking Matter? Wenyan as a Compact Reasoning Register for Chinese LLMs

文言思考：古文提升大语言模型推理效率

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-15 04:00

研究人员提出了一种名为“文言思考”的新方法，用于中文大语言模型（LLMs），该方法在中间推理步骤中使用古文（文言文）。该方法旨在通过缩短推理痕迹的长度来提高效率，因为现代白话中文的推理痕迹通常冗长。通过在用古文重写的推理痕迹上微调Qwen3-4B-Thinking-2507模型，研究人员在MMLU-ProX-Lite-ZH、CEVAL和AGIEval-CN等多个基准测试中展示了效率的显著提升，同时保持了有竞争力的性能。 AI

影响这项研究表明，语言的语域是提高中文大语言模型可见推理效率的一个实用设计轴。

排序理由该集群包含一篇学术论文，详细介绍了一种新颖的大语言模型推理方法。[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: 14 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇学术论文，详细介绍了一种新颖的大语言模型推理方法。[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: 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]

arXiv cs.CL TIER_1 English(EN) · Minghan Li, Guodong Zhou · 2026-10-15 04:00

思维的语言风格是否重要？文言作为中文大模型的紧凑推理语域

arXiv:2610.19741v1 Announce Type: new Abstract: Visible reasoning traces of Chinese reasoning-oriented LLMs are typically written in modern vernacular Chinese, which may be verbose under limited decoding budgets. We ask whether the linguistic register of reasoning can be used to …

本文如何被排名

报道来源 [1]

思维的语言风格是否重要？文言作为中文大模型的紧凑推理语域

相关实体

相关话题