Rethinking Multilingual Continual Pretraining: Data Mixing for Adapting LLMs Across Languages and Resources

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Li, Zihao, Ji, Shaoxiong, Luo, Hengyu, Tiedemann, Jörg
Format:	Preprint
Published:	2025
Subjects:	Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Internet

https://arxiv.org/abs/2504.04152