Abstract
Machine Reading Comprehension (MRC) is one of the basic tasks in Natural Language Processing, which can be used to test the ability of natural language comprehension of a model. Several models and methods have succeeded in MRC, but they are seldom robust enough, which makes them work ill in the real-world question and answering tasks filled up with noises and confusions. To tackle with that, we propose a model based on pretrained Chinese-MacBERT-Large. In order to get enough robustness, we apply two adversarial methods, i.e. FGM & VAT in training process. The experiments on DuReader_robust show that our model outperforms the baseline based on Chinese-MacBERT-Base without adversarial methods, which means our model is more stable to noises and confusions.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
