Turn Waste into Worth: Rectifying Top-k Router of MoE Zhiyuan Zeng author Qipeng Guo author Zhaoye Fei author Zhangyue Yin author Yunhua Zhou author Linyang Li author Tianxiang Sun author Hang Yan author Dahua Lin author Xipeng Qiu author 2024-11 text Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing Yaser Al-Onaizan editor Mohit Bansal editor Yun-Nung Chen editor Association for Computational Linguistics Miami, Florida, USA conference publication zeng-etal-2024-turn 10.18653/v1/2024.emnlp-main.739 https://aclanthology.org/2024.emnlp-main.739/ 2024-11 13305 13320