NEZHA: A Zero-sacrifice and Hyperspeed Decoding Architecture for Generative Recommendations.

@inproceedings{wang-2026-nezha, author = {Yejing Wang and Shengyu Zhou and Jinyu Lu and Ziwei Liu and Langming Liu and Maolin Wang and Wenlin Zhang and Feng Li and Wenbo Su and Pengjie Wang and Jian Xu and Xiangyu Zhao}, editor = {Hakim Hacid and Yoelle Maarek and Francesco Bonchi and Ido Guy and Emine Yilmaz}, title = {{NEZHA: A Zero-sacrifice and Hyperspeed Decoding Architecture for Generative Recommendations}}, booktitle = {{Proceedings of the ACM Web Conference 2026, WWW 2026, Dubai, United Arab Emirates, originally scheduled for April 13-17, 2026, rescheduled for June 29 - July 3, 2026}}, pages = {8073--8082}, publisher = {ACM}, year = {2026}, url = {https://doi.org/10.1145/3774904.3792797}, doi = {https://doi.org/10.1145/3774904.3792797} }