Two Pathways to Truthfulness: On the Intrinsic Encoding of LLM Hallucinations

arXiv:2601.07422v2 Announce Type: replace-cross Abstract: Despite their impressive capabilities, large language models (LLMs) frequently generate hallucinations. Previous work shows that their internal states encode rich signals of truthfulness, yet the origins and mechanisms of these signals remain unclear. In this paper, we demonstrate that truthfulness cues arise from two distinct information pathways: (1) a Question-Anchored pathway that depends on question-answer information flow, and (2) a