ERNIE: Enhanced Representation through Knowledge Integration

19 Apr 2019  ·  Yu Sun, Shuohuan Wang, Yukun Li, Shikun Feng, Xuyi Chen, Han Zhang, Xin Tian, Danxiang Zhu, Hao Tian, Hua Wu ·

We present a novel language representation model enhanced by knowledge called ERNIE (Enhanced Representation through kNowledge IntEgration). Inspired by the masking strategy of BERT, ERNIE is designed to learn language representation enhanced by knowledge masking strategies, which includes entity-level masking and phrase-level masking. Entity-level strategy masks entities which are usually composed of multiple words.Phrase-level strategy masks the whole phrase which is composed of several words standing together as a conceptual unit.Experimental results show that ERNIE outperforms other baseline methods, achieving new state-of-the-art results on five Chinese natural language processing tasks including natural language inference, semantic similarity, named entity recognition, sentiment analysis and question answering. We also demonstrate that ERNIE has more powerful knowledge inference capacity on a cloze test.

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Chinese Sentiment Analysis ChnSentiCorp ERNIE Accuracy 95.4 # 3
Chinese Sentiment Analysis ChnSentiCorp Dev ERNIE Accuracy 95.2 # 3
Chinese Sentence Pair Classification LCQMC ERNIE Accuracy 87.4 # 4
Chinese Sentence Pair Classification LCQMC Dev ERNIE Accuracy 89.7 # 3
Chinese Named Entity Recognition MSRA ERNIE F1 93.8 # 14
Chinese Named Entity Recognition MSRA Dev ERNIE F1 95 # 3
Chinese Sentence Pair Classification NLPCC-DBQA ERNIE MRR 95.1 # 3
F1 82.7 # 2
Chinese Sentence Pair Classification NLPCC-DBQA Dev ERNIE MRR 95 # 3
F1 82.3 # 1
Natural Language Inference XNLI Chinese ERNIE Accuracy 78.4 # 3
Natural Language Inference XNLI Chinese Dev ERNIE Accuracy 79.9 # 3

Methods