Bitfit: Simple Parameter-efficient Fine-tuning For Transformer-based Masked Language-models · The Large Language Model Bible Contribute to LLM-Bible

Bitfit: Simple Parameter-efficient Fine-tuning For Transformer-based Masked Language-models

Zaken Elad Ben, Ravfogel Shauli, Goldberg Yoav. Arxiv 2021

[Paper]    
BERT Ethics And Bias Fine Tuning Model Architecture Pretraining Methods Reinforcement Learning Training Techniques Transformer

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.

Similar Work