Interrupting the Loop: Periodic Subject Changes Raise Judged Surprise and Connection in Base Language Models

arXiv:2608.19893v2 Announce Type: replace-cross Abstract: Where does the novelty a base language model produces with no task come from, and what can an LLM judge of a long stream actually see? We dismantle a cognitively inspired generation loop…

Read the full story at arXiv cs.AI ↗