Fine-tuning BERT for classification with additional variables

Question

I\'m fine-tuning BERT for classification task using HF\'s transformers (transformers.BertForSequenceClassification to be precise).

I am wondering if there is