[Retracted] Evidence Prediction Method Based on Sentence Selection for Legal Documents

Table 4

RMSE scores of models before and after feature fusion.

	CNN	LSTM	GRU	TextCNN	Causal textCNN
Before fusion	30.23	26.03	40.81	23.87	28.90
After fusion	25.23	33.03	32.75	29.87	18.72