notesum.ai

Published at November 25

BayLing 2: A Multilingual Large Language Model with Efficient Language Alignment

cs.CL

cs.AI

Released Date: November 25, 2024

Authors: Shaolei Zhang¹, Kehao Zhang¹, Qingkai Fang¹, Shoutao Guo¹, Yan Zhou¹, Xiaodong Liu², Yang Feng³

Aff.: ¹Key Laboratory of Intelligent Information Processing, Institute of Computing Technology, Chinese Academy of Sciences (ICT/CAS); University of Chinese Academy of Sciences, Beijing, China; ²Research Center Of Distributed Systems, Institute of Computing Technology, Chinese Academy of Sciences (ICT/CAS); ³Key Laboratory of Intelligent Information Processing, Institute of Computing Technology, Chinese Academy of Sciences (ICT/CAS); Key Laboratory of AI Safety, Chinese Academy of Sciences; University of Chinese Academy of Sciences, Beijing, China

Arxiv: http://arxiv.org/abs/2411.16300v1

Models	X $\Rightarrow$ English		English $\Rightarrow$ X		X $\Rightarrow$ Chinese		Chinese $\Rightarrow$ X
Models	BLEU	COMET	BLEU	COMET	BLEU	COMET	BLEU	COMET
Llama-1-7B	14.07	60.94	6.93	49.73	0.93	40.88	1.85	44.88
BayLing-1-7B	14.70	61.93	7.04	49.33	1.58	46.22	1.56	48.78
Llama-2-7B-Chat	15.39	63.95	7.45	50.97	1.75	47.19	1.57	45.70
BayLing-2-7B	17.71	67.15	8.02	52.37	2.70	51.44	2.32	49.37
Llama-3-8B-Instruct	25.20	76.60	16.59	67.17	11.79	71.91	8.95	63.57
BayLing-3-8B	26.77	77.03	17.91	70.88	11.31	69.43	10.64	67.86