Crossing Linguistic Horizons

Finetuning and Comprehensive Evaluation of Vietnamese Large Language Models

Few-Shot Knowledge Leaderboard

Models ZaloE2E ViMMRC
EM F1 AC F1 AR ECE A@10
URA-LLaMa 70B 0.34 ± 0.02 0.50 ± 0.02 0.78 ± 0.02 0.63 ± 0.03 0.90 ± 0.01 0.13 ± 0.02 0.96 ± 0.03
URA-LLaMa 13B 0.26 ± 0.02 0.40 ± 0.02 0.62 ± 0.02 0.50 ± 0.02 0.69 ± 0.02 0.18 ± 0.02 0.65 ± 0.07
URA-LLaMa 7B 0.14 ± 0.02 0.25 ± 0.02 0.42 ± 0.02 0.33 ± 0.02 0.61 ± 0.02 0.13 ± 0.02 0.39 ± 0.07
LLaMa-2 13B 0.22 ± 0.02 0.36 ± 0.02 0.58 ± 0.02 0.46 ± 0.02 0.62 ± 0.02 0.28 ± 0.02 0.77 ± 0.06
LLaMa-2 7B 0.07 ± 0.01 0.15 ± 0.01 0.30 ± 0.02 0.23 ± 0.02 0.56 ± 0.02 0.43 ± 0.02 0.16 ± 0.05
Vietcuna 7B 0.07 ± 0.01 0.19 ± 0.01 0.31 ± 0.02 0.18 ± 0.01 0.50 ± 0.00 0.06 ± 0.02 0.31 ± 0.06
MixSUra 8x7B 0.19 ± - 0.34 ± - 0.65 ± - 0.64 ± - 0.54 ± - 0.29 ± - 0.65 ± -
GPT-3.5 0.49 ± 0.02 0.64 ± 0.02 0.90 ± 0.01 0.73 ± 0.03 - 0.66 ± 0.01 0.91 ± 0.04
GPT-4 0.49 ± 0.02 0.64 ± 0.02 0.91 ± 0.01 0.73 ± 0.04 - 0.66 ± 0.01 0.91 ± 0.04