English(EN) "Not" is Not Lost in Translation: A Causal Analysis of Negation Representations in Multilingual LLMs

多语言大语言模型在不同语言中展现出共享的否定表达

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-26 04:00

研究人员调查了多语言大语言模型（LLMs）如何在不同语言中表示否定。他们的研究利用了对合成和翻译数据集的因果干预，发现这些模型在很大程度上共享否定的底层表达。稀疏自编码器中的单个特征足以控制跨语言的极性，表明否定是以稀疏特征表示的，并且可以被因果地操纵。 AI

影响这项研究表明，大语言模型可能会发展出更强大的跨语言理解能力，从而可能改进翻译和多语言应用。

排序理由该集群包含一篇详细介绍大语言模型能力研究结果的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

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: 16 / 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大语言模型能力研究结果的学术论文。[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, 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。

arXiv cs.AI TIER_1 English(EN) · Giang Son Nguyen, Wray Buntine, Dung D. Le, Wenya Wang · 2026-10-26 04:00

“不”字翻译未丢失：多语言大语言模型否定表征的因果分析

arXiv:2610.32281v1 Announce Type: cross Abstract: Negation occurs in every language, yet its surface realization differs drastically. Despite this, multilingual large language models (LLMs) translate negation robustly, raising the question of whether they share an underlying repr…