CARO: Chain-of-Analogy Reasoning Optimization for Robust Content Moderation.

Bingzhe Wu, Haotian Lu, Yuchen Mou

Browse the full ACL paper archive.