Towards More Equitable Question Answering Systems: How Much More Data Do You Need?

Question answering (QA) in English has been widely explored, but multilingual\ndatasets are relatively new, with several methods attempting to bridge the gap\nbetween high- and low-resourced languages using data augmentation through\ntranslation and cross-lingual transfer. In this project, we take a step back\nand study which approaches allow us to take the most advantage of existing\nresources in order to produce QA systems in many languages. Specifically, we\nperform extensive analysis to measure the efficacy of few-shot approaches\naugmented with automatic translations and permutations of\ncontext-question-answer pairs. In addition, we make suggestions for future\ndataset development efforts that make better use of a fixed annotation budget,\nwith a goal of increasing the language coverage of QA datasets and systems.\nCode and data for reproducing our experiments are available here:\nhttps://github.com/NavidRajabi/EMQA.\n