Mitigate Position Bias in Large Language Models via Scaling a Single Dimension

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Yu, Yijiong, Jiang, Huiqiang, Luo, Xufang, Wu, Qianhui, Lin, Chin-Yew, Li, Dongsheng, Yang, Yuqing, Huang, Yongfeng, Qiu, Lili
Format:	Preprint
Published:	2024
Subjects:	Computation and Language Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Internet

https://arxiv.org/abs/2406.02536