Pair-In, Pair-Out: Latent Multi-Token Prediction for Efficient LLMs 文章

ArXiv CS.CL2026-06-01NEWSen作者: Wenhui Tan, Minghao Li, Xiaoqian Ma, Siqi Fan, Xiusheng Huang, Liujie Zhang, Ruihua Song, Weihang Chen

Pair-In, Pair-Out: Latent Multi-Token Prediction for Efficient LLMs · 相关技术