When to Fold'em: How to answer Unanswerable questions
This addresses the challenge of handling unanswerable questions in QA systems, though it appears incremental as it builds on existing models and datasets.
The paper tackles the problem of answering unanswerable questions by fine-tuning pre-trained question-answering models on the SQuAD2.0 dataset, achieving a 2% point improvement in F1 score with reduced training time.
We present 3 different question-answering models trained on the SQuAD2.0 dataset -- BIDAF, DocumentQA and ALBERT Retro-Reader -- demonstrating the improvement of language models in the past three years. Through our research in fine-tuning pre-trained models for question-answering, we developed a novel approach capable of achieving a 2% point improvement in SQuAD2.0 F1 in reduced training time. Our method of re-initializing select layers of a parameter-shared language model is simple yet empirically powerful.