TY - JOUR
T1 - Unsupervised Question Answering via Answer Diversifying
AU - Nie, Yuxiang
AU - Huang, Heyan
AU - Chi, Zewen
AU - Mao, Xian Ling
N1 - Publisher Copyright:
© 2022 Proceedings - International Conference on Computational Linguistics, COLING. All rights reserved.
PY - 2022
Y1 - 2022
N2 - Unsupervised question answering is an attractive task due to its independence on labeled data. Previous works usually make use of heuristic rules as well as pre-trained models to construct data and train QA models. However, most of these works regard named entity (NE) as the only answer type, which ignores the high diversity of answers in the real world. To tackle this problem, we propose a novel unsupervised method by diversifying answers, named DiverseQA. Specifically, the proposed method is composed of three modules: data construction, data augmentation and denoising filter. Firstly, the data construction module extends the extracted named entity into a longer sentence constituent as the new answer span to construct a QA dataset with diverse answers. Secondly, the data augmentation module adopts an answer-type dependent data augmentation process via adversarial training in the embedding level. Thirdly, the denoising filter module is designed to alleviate the noise in the constructed data. Extensive experiments show that the proposed method outperforms previous unsupervised models on five benchmark datasets, including SQuADv1.1, NewsQA, TriviaQA, BioASQ, and DuoRC. Besides, the proposed method shows strong performance in the few-shot learning setting.
AB - Unsupervised question answering is an attractive task due to its independence on labeled data. Previous works usually make use of heuristic rules as well as pre-trained models to construct data and train QA models. However, most of these works regard named entity (NE) as the only answer type, which ignores the high diversity of answers in the real world. To tackle this problem, we propose a novel unsupervised method by diversifying answers, named DiverseQA. Specifically, the proposed method is composed of three modules: data construction, data augmentation and denoising filter. Firstly, the data construction module extends the extracted named entity into a longer sentence constituent as the new answer span to construct a QA dataset with diverse answers. Secondly, the data augmentation module adopts an answer-type dependent data augmentation process via adversarial training in the embedding level. Thirdly, the denoising filter module is designed to alleviate the noise in the constructed data. Extensive experiments show that the proposed method outperforms previous unsupervised models on five benchmark datasets, including SQuADv1.1, NewsQA, TriviaQA, BioASQ, and DuoRC. Besides, the proposed method shows strong performance in the few-shot learning setting.
UR - http://www.scopus.com/inward/record.url?scp=85161146728&partnerID=8YFLogxK
M3 - Conference article
AN - SCOPUS:85161146728
SN - 2951-2093
VL - 29
SP - 1732
EP - 1742
JO - Proceedings - International Conference on Computational Linguistics, COLING
JF - Proceedings - International Conference on Computational Linguistics, COLING
IS - 1
T2 - 29th International Conference on Computational Linguistics, COLING 2022
Y2 - 12 October 2022 through 17 October 2022
ER -