Table 4 Comparison of the performance of different models in the language generation task of Yongju.

From: Automatic generation and transfer of Yongju opera language using AIGC technology and style mapping

Model

Standard deviation

Standard error

Sample size

p value

TFCPG vs. transformer

20.175

2.5

500

0.000***

TFCPG vs. PLM

20.225

2.6

500

0.000***

TFCPG vs. GPT

19.95

2.4

500

0.0008**

TFCPG vs. BERT

15.825

2.2

500

0.016**

  1. *p < 0.1, **p < 0.05, ***p < 0.01.
Back to article page