Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

< Back to Article

Fig 1.

The transformer structure used in PLM.

More »

Fig 1 Expand

Fig 2.

Classic knowledge distillation approach.

More »

Fig 2 Expand

Fig 3.

The structure of BERTtoCNN model.

More »

Fig 3 Expand

Fig 4.

Text-CNN as the student model.

More »

Fig 4 Expand

Table 1.

The EDA result of Climate Change is Concern.

More »

Table 1 Expand

Table 2.

The SemEval-2016 stance dataset.

More »

Table 2 Expand

Table 3.

NLPCC-ICCPOL-2016 stance dataset.

More »

Table 3 Expand

Table 4.

The performance of BERTtoCNN compared with the baseline methods.

More »

Table 4 Expand

Fig 5.

F-score curves with the variation of parameter α and T.

Among them, α is the parameter of balancing the soft and hard labels in the knowledge distillation loss, and T is the temperature, which is used to adjust the smoothness of prediction distribution in knowledge distillation.

More »

Fig 5 Expand

Table 5.

F-score with the variation of hyperparameter γ.

More »

Table 5 Expand

Table 6.

The comparison of EDA data and original data on the model.

Within, we set n_aug = 1 represents the result of the original data without EDA on the model.

More »

Table 6 Expand