BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
This work provides a foundational model for natural language processing, enabling broad improvements in tasks like question answering and inference without task-specific architectures.
BERT tackled the problem of language understanding by pre-training deep bidirectional representations from unlabeled text, achieving state-of-the-art results on eleven NLP tasks, such as an 80.5% GLUE score and 93.2 F1 on SQuAD v1.1.
We introduce a new language representation model called BERT, which stands for Bidirectional Encoder Representations from Transformers. Unlike recent language representation models, BERT is designed to pre-train deep bidirectional representations from unlabeled text by jointly conditioning on both left and right context in all layers. As a result, the pre-trained BERT model can be fine-tuned with just one additional output layer to create state-of-the-art models for a wide range of tasks, such as question answering and language inference, without substantial task-specific architecture modifications. BERT is conceptually simple and empirically powerful. It obtains new state-of-the-art results on eleven natural language processing tasks, including pushing the GLUE score to 80.5% (7.7% point absolute improvement), MultiNLI accuracy to 86.7% (4.6% absolute improvement), SQuAD v1.1 question answering Test F1 to 93.2 (1.5 point absolute improvement) and SQuAD v2.0 Test F1 to 83.1 (5.1 point absolute improvement).