English(EN) The Geometry of Ignorance: LLMs Know When to Temper Bayesian Priors

LLM使用“无知方向”来调整贝叶斯先验

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

研究人员在大型语言模型（LLM）中发现了一种几何特性，该特性量化了模型在面对有限上下文时对先验知识的依赖程度。这种“无知方向”被编码在反嵌入矩阵中，并充当贝叶斯先验，模型在不确定时会默认使用它。在Llama、Qwen、Gemma和Pythia等各种模型系列中都观察到了这种现象，与其参数大小无关。 AI

影响这项研究为理解和潜在控制LLM的不确定性和校准提供了一个新的视角。

排序理由该集群包含一篇详细介绍LLM行为新发现的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv stat.ML 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: 33 / 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LLM行为新发现的学术论文。[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。

报道来源 [1]

arXiv stat.ML TIER_1 English(EN) · Toni J. B. Liu, Jiajun Bao, Yizhou Liu, Gurbir Arora, Nicolas Boull\'e, Rapha\"el Sarfati, Christopher J. Earls · 2026-09-04 04:00

无知的几何学：LLM知道何时调整贝叶斯先验

arXiv:2609.02959v1 Announce Type: cross Abstract: What does a language model predict when it has few clues? The answer lurks in its unembedding geometry: a single direction of the unembedding matrix encodes the unigram distribution of the training corpus, which serves as the Baye…

本文如何被排名

报道来源 [1]

无知的几何学：LLM知道何时调整贝叶斯先验

相关实体

相关话题