Two papers about reducing hallucination in dialogue system and debiased natural language understanding were accepted by AAAI’23.