Rare Tokens Degenerate All Tokens: Improving Neural Text Generation via Adaptive Gradient Gating for Rare Token Embeddings

Rare Tokens Degenerate All Tokens: Improving Neural Text Generation via Adaptive Gradient Gating for Rare Token Embeddings Sangwon Yu author Jongyoon Song author Heeseung Kim author Seongmin Lee author Woo-Jong Ryu author Sungroh Yoon author 2022-05 text Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) Smaranda Muresan editor Preslav Nakov editor Aline Villavicencio editor Association for Computational Linguistics Dublin, Ireland conference publication yu-etal-2022-rare 10.18653/v1/2022.acl-long.3 https://aclanthology.org/2022.acl-long.3/ 2022-05 29 45