Generative Data Augmentation using LLMs improves Distributional Robustness in Question Answering
This work addresses domain generalization for question answering models, which is an incremental improvement in handling unknown target domains.
The paper tackled the problem of improving distributional robustness in question answering models under natural distribution shifts by using generative data augmentation with LLMs, and demonstrated that augmenting datasets with generated contexts and QA pairs leads to better robustness across four datasets.
Robustness in Natural Language Processing continues to be a pertinent issue, where state of the art models under-perform under naturally shifted distributions. In the context of Question Answering, work on domain adaptation methods continues to be a growing body of research. However, very little attention has been given to the notion of domain generalization under natural distribution shifts, where the target domain is unknown. With drastic improvements in the quality and access to generative models, we answer the question: How do generated datasets influence the performance of QA models under natural distribution shifts? We perform experiments on 4 different datasets under varying amounts of distribution shift, and analyze how "in-the-wild" generation can help achieve domain generalization. We take a two-step generation approach, generating both contexts and QA pairs to augment existing datasets. Through our experiments, we demonstrate how augmenting reading comprehension datasets with generated data leads to better robustness towards natural distribution shifts.