A BERT Baseline for the Natural Questions

24 Jan 2019  ·  Chris Alberti, Kenton Lee, Michael Collins ·

This technical note describes a new baseline for the Natural Questions. Our model is based on BERT and reduces the gap between the model F1 scores reported in the original dataset paper and the human upper bound by 30% and 50% relative for the long and short answer tasks respectively. This baseline has been submitted to the official NQ leaderboard at ai.google.com/research/NaturalQuestions. Code, preprocessed data and pretrained model are available at https://github.com/google-research/language/tree/master/language/question_answering/bert_joint.

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Question Answering Natural Questions (long) BERTjoint F1 64.7 # 6