Generalizing Question Answering System with Pre-trained Language Model Fine-tuning

Citations Over TimeTop 10% of 2019 papers

Abstract

With a large number of datasets being released and new techniques being proposed, Question answering (QA) systems have witnessed great breakthroughs in reading comprehension (RC)tasks. However, most existing methods focus on improving in-domain performance, leaving open the research question of how these mod-els and techniques can generalize to out-of-domain and unseen RC tasks. To enhance the generalization ability, we propose a multi-task learning framework that learns the shared representation across different tasks. Our model is built on top of a large pre-trained language model, such as XLNet, and then fine-tuned on multiple RC datasets. Experimental results show the effectiveness of our methods, with an average Exact Match score of 56.59 and an average F1 score of 68.98, which significantly improves the BERT-Large baseline by8.39 and 7.22, respectively

Related Papers

• → A strong baseline for question relevancy ranking(2018)9 cited
• → SemEval-2019 Task 8: Fact Checking in Community Question Answering Forums(2019)9 cited
• → SemEval-2019 Task 8: Fact Checking in Community Question Answering\n Forums(2019)1 cited
• → A Question Answer Approach to Building Semantic Memory(2012)