Sangwon Yu, Jongyoon Song, Heeseung Kim, Seongmin Lee 0005, Woo-Jong Ryu, Sungroh Yoon. Rare Tokens Degenerate All Tokens: Improving Neural Text Generation via Adaptive Gradient Gating for Rare Token Embeddings. In Smaranda Muresan, Preslav Nakov, Aline Villavicencio, editors, Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), ACL 2022, Dublin, Ireland, May 22-27, 2022. pages 29-45, Association for Computational Linguistics, 2022. [doi]
Abstract is missing.