SOURCE-LINKED INTELLIGENCE

Representational alignment yields generalizable safety in language models

arXiv · AI, language, vision and robotics · article · Sep 3, 2026 · UTC

Aligning large language models (LLMs) is essential for their safe deployment. Current alignment methods mainly optimize observable responses, yet models remain vulnerable when the same harmful intent is recast in unfamiliar or adversarial forms that humans can easily recognize. Prototype theory offers an account of this adaptability. Human concepts are represented around central cases, and new instances are categorized according to their graded typicality relative to these prototypes. Here we show that such categorization of moral concepts is weakly preserved in current LLMs. Across 23 LLMs, m

Read original source ↗ Open in workspace

recordType

paper

region

Global