When in Doubt, Ask: Generating Answerable and Unanswerable Questions, Unsupervised
This addresses the data scarcity issue for QA model developers, but it is incremental as it builds on existing methods for synthetic data generation.
The paper tackled the problem of costly human-generated training data for Question Answering (QA) by augmenting a human-made dataset with synthetic answerable and unanswerable questions, resulting in tangible performance improvements, with F1 score gains of up to 6.7%.
Question Answering (QA) is key for making possible a robust communication between human and machine. Modern language models used for QA have surpassed the human-performance in several essential tasks; however, these models require large amounts of human-generated training data which are costly and time-consuming to create. This paper studies augmenting human-made datasets with synthetic data as a way of surmounting this problem. A state-of-the-art model based on deep transformers is used to inspect the impact of using synthetic answerable and unanswerable questions to complement a well-known human-made dataset. The results indicate a tangible improvement in the performance of the language model (measured in terms of F1 and EM scores) trained on the mixed dataset. Specifically, unanswerable question-answers prove more effective in boosting the model: the F1 score gain from adding to the original dataset the answerable, unanswerable, and combined question-answers were 1.3%, 5.0%, and 6.7%, respectively. [Link to the Github repository: https://github.com/lnikolenko/EQA]