LayerCake: Token-Aware Contrastive Decoding within Large Language Model Layers

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Zhu, Jingze, Wu, Yongliang, Zhu, Wenbo, Cao, Jiawang, Zheng, Yanqiang, Chen, Jiawei, Yang, Xu, Schiele, Bernt, Fischer, Jonas, Hu, Xinting
Format:	Preprint
Published:	2025
Subjects:	Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Internet

https://arxiv.org/abs/2507.04404