Reinforcement Learning for Hallucination-Aware Reasoning and Factuality Calibration in Large Language Models. JDIAS 2026, 1 (3).