Denoising Distantly Supervised Open-Domain Question Answering

ACL 2018 Yankai LinHaozhe JiZhiyuan LiuMaosong Sun

Distantly supervised open-domain question answering (DS-QA) aims to find answers in collections of unlabeled text. Existing DS-QA models usually retrieve related paragraphs from a large-scale corpus and apply reading comprehension technique to extract answers from the most relevant paragraph... (read more)

PDF Abstract

Evaluation results from the paper


Task Dataset Model Metric name Metric value Global rank Compare
Open-Domain Question Answering Quasar Denoising QA EM (Quasar-T) 42.2 # 2
Open-Domain Question Answering Quasar Denoising QA F1 (Quasar-T) 49.3 # 2
Open-Domain Question Answering SearchQA Denoising QA Unigram Acc - # 5
Open-Domain Question Answering SearchQA Denoising QA N-gram F1 - # 6
Open-Domain Question Answering SearchQA Denoising QA EM 58.8 # 1
Open-Domain Question Answering SearchQA Denoising QA F1 64.5 # 2