Reduce, Reuse, Recycle: New uses for old QA resources

TL;DR

This paper demonstrates that repurposing generic QA datasets like SQuAD for relation extraction tasks enhances zero-shot performance and robustness, using standard QA models without modifications.

Contribution

It introduces a novel approach of applying generic QA data and models to relation extraction, showing improved generalization and simplicity.

Findings

SQuAD-based training outperforms task-specific data in zero-shot settings.

Standard QA architectures can be directly applied to slot filling queries.

Repurposing QA resources simplifies relation extraction tasks.

Abstract

We investigate applying repurposed generic QA data and models to a recently proposed relation extraction task. We find that training on SQuAD produces better zero-shot performance and more robust generalisation compared to the task specific training set. We also show that standard QA architectures (e.g. FastQA or BiDAF) can be applied to the slot filling queries without the need for model modification.