AI RESEARCH
When Corrective Hints Hurt: Prompt Design in Reasoner-Guided Repair of LLM Overcaution on Entailed Negations under OWL~2~DL
arXiv CS.AI
•
ArXi:2604.23398v1 Announce Type: new We report a reproducible error pattern in GPT-5.4 on OWL~2~DL compliance queries: the model frequently answers ``unknown'' when the reasoner-entailed answer is ``no'' under \emph{FunctionalProperty} closure or class \emph{disjointness