arXivKaishen Wang, Tong Zheng, Xuehao Cui, Ruibo Chen, Tianyi Xiong, Heng HuangMon, Jul 6, 2026, 10:34 PM PDT
score 16.6
New method stops AI reasoning from inventing false facts
Original: Mitigating Factual Hallucination in Large Reasoning Models via Mixed-Mode Advantage Regularization
Source: arxiv.org ↗
Writing ELI5 summary…