2019 RoBERTaARobustlyOptimizedBERTPr

From GM-RKB
(Redirected from Liu, Ott et al., 2019)
Jump to navigation Jump to search

Subject Headings: RoBERTa System, RoBERTa Model, BERT.

Notes

Cited By

Quotes

Abstract

Language model pretraining has led to significant performance gains but careful comparison between different approaches is challenging. Training is computationally expensive, often done on private datasets of different sizes, and, as we will show, hyperparameter choices have significant impact on the final results. We present a replication study of BERT pretraining (Devlin et al., 2019) that carefully measures the impact of many key hyperparameters and training data size. We find that BERT was significantly undertrained, and can match or exceed the performance of every model published after it. Our best model achieves state-of-the-art results on GLUE, RACE and SQuAD. These results highlight the importance of previously overlooked design choices, and raise questions about the source of recently reported improvements. We release our models and code.

1 Introduction

2 Background

3 Experimental Setup

4 Training Procedure Analysis

5 RoBERTa

6 Related Work

7 Conclusion

References

;

 AuthorvolumeDate ValuetitletypejournaltitleUrldoinoteyear
2019 RoBERTaARobustlyOptimizedBERTPrOmer Levy
Luke Zettlemoyer
Danqi Chen
Veselin Stoyanov
Yinhan Liu
Myle Ott
Naman Goyal
Jingfei Du
Mandar Joshi
Mike Lewis
RoBERTa: {A} Robustly Optimized {BERT} Pretraining Approach2019