BitFit: Simple Parameter-efficient Fine-tuning for Transformer-based Masked Language-models
This provides a simple, efficient fine-tuning approach for NLP practitioners, though it is incremental as it builds on existing sparse fine-tuning ideas.
The authors tackled the problem of parameter-efficient fine-tuning for transformer-based masked language models by introducing BitFit, a method that modifies only bias terms, and showed it is competitive with full fine-tuning on small-to-medium data and with other sparse methods on larger data.
We introduce BitFit, a sparse-finetuning method where only the bias-terms of the model (or a subset of them) are being modified. We show that with small-to-medium training data, applying BitFit on pre-trained BERT models is competitive with (and sometimes better than) fine-tuning the entire model. For larger data, the method is competitive with other sparse fine-tuning methods. Besides their practical utility, these findings are relevant for the question of understanding the commonly-used process of finetuning: they support the hypothesis that finetuning is mainly about exposing knowledge induced by language-modeling training, rather than learning new task-specific linguistic knowledge.