English(EN) Inoculation Midtraining with Learned Neologisms

新的中间训练接种技术塑造LLM行为

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-15 04:00

研究人员开发了一种称为中间训练接种的新技术来塑造大型语言模型（LLM）的行为。该方法在早期训练阶段引入了一个特殊的``，将不良行为与该标记关联起来。然后，模型仅在该特定上下文中进行不安全数据的训练。当在上下文之外进行评估时，模型显示出失调的减少，同时保留了说不同语言等有益的特性。然而，该方法的有效性对训练配置敏感，并且不总是优于标准的接种提示，这表明其在人工智能安全框架中的实际应用还需要进一步研究。 AI

影响这项研究探索了一种通过隔离不良行为来提高LLM安全性的新方法，尽管其实际应用需要进一步发展。

排序理由该集群包含一篇详细介绍LLM训练新方法的论文。[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 →

本文如何被排名

Signal score: 15 / 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训练新方法的论文。[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, safety; 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) · Kyle O'Brien, Edward James Young, Puria Radmard, Nathalie Kirch, Cameron Tice, Tomek Korbak, David Demitri Africa · 2026-09-15 04:00

在学习新词的中间训练中接种

arXiv:2609.15886v1 Announce Type: new Abstract: Large language models (LLMs) often learn both desirable and undesirable properties during post-training. We study whether midtraining, an earlier training stage, can shape which of these properties later generalise. We introduce Ino…

本文如何被排名

报道来源 [1]

在学习新词的中间训练中接种

相关实体

相关话题