Shallow Prefill, Deep Decoding: Efficient Long-Context Inference via Layer-Asymmetric KV Visibility

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Oh, Jungsuk, Jeon, Hyeseo, Ji, Hyunjune, Kong, Kyongmin, Lee, Jay-Yoon
Format:	Preprint
Published:	2026
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/2605.06105