arXiv cs.AI / cs.LG / cs.CL·16d agoDomain-Specific Hallucination Detection in Large Language Models#calibration#deberta-v3#dpoAI research1