ExtremeBERT: A Toolkit for Accelerating Pretraining of Customized BERT
This toolkit makes BERT pretraining more affordable for researchers and industry with limited resources, though it is incremental as it builds on existing methods.
The paper tackles the problem of expensive BERT pretraining by introducing ExtremeBERT, a toolkit that accelerates and customizes pretraining, achieving over 6x and 9x time reduction for BERT Base and Large respectively while matching or improving GLUE scores.
In this paper, we present ExtremeBERT, a toolkit for accelerating and customizing BERT pretraining. Our goal is to provide an easy-to-use BERT pretraining toolkit for the research community and industry. Thus, the pretraining of popular language models on customized datasets is affordable with limited resources. Experiments show that, to achieve the same or better GLUE scores, the time cost of our toolkit is over $6\times$ times less for BERT Base and $9\times$ times less for BERT Large when compared with the original BERT paper. The documentation and code are released at https://github.com/extreme-bert/extreme-bert under the Apache-2.0 license.