MIT Press

Skip to Main Content

Table 9

The BERT-A-T model hyperparameters. H is defined as in Table 8.

Layer	Input Dimensions	Output Dimensions
BERT Pretrained Encoder	Interview text	H × # Q-A pairs
Attention	H × Max # Q-A pairs in batch	H
Linear Output	H	1

Layer	Input Dimensions	Output Dimensions
BERT Pretrained Encoder	Interview text	H × # Q-A pairs
Attention	H × Max # Q-A pairs in batch	H
Linear Output	H	1