BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Jacob Devlin, Ming-Wei Chang, Kenton Lee, Kristina Toutanova
Code Available — Be the first to reproduce this paper.
ReproduceCode
- github.com/google-research/bertOfficialIn papertf★ 39,935
- github.com/huggingface/transformerspytorch★ 158,292
- github.com/labmlai/annotated_deep_learning_paper_implementationspytorch★ 66,103
- github.com/graykode/nlp-tutorialpytorch★ 14,880
- github.com/PaddlePaddle/PaddleNLPpaddle★ 12,937
- github.com/PaddlePaddle/modelspaddle★ 6,946
- github.com/shreyashankar/gpt3-sandboxnone★ 2,880
- github.com/alibaba/EasyNLPjax★ 2,181
- github.com/yoshitomo-matsubara/torchdistillpytorch★ 1,603
- github.com/lukemelas/PyTorch-Pretrained-ViTpytorch★ 853
Abstract
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).