Ablating Alignment Signatures

Measuring, Localizing, and Ablating Alignment Signatures in LLMs

arXiv:2605.30526v1 Announce Type: new Abstract: Aligned language models often exhibit a recognizable AI-like style, yet its connection to post-training and internal representations remains poorly understood. In this work, we study whether post-training introduces or amplifies AI-like stylistic regularities and whether these regularities have a localized internal signature. To this end, we compare human text, base-model generations, and aligned-model generations under matched human-source...