MemSum-DQA: Adapting An Efficient Long Document Extractive Summarizer for Document Question Answering
Published in arXiv preprint arXiv:2310.06436, 2024
Recommended citation: Gu, Nianlong, Yingqiang Gao, and Richard Hahnloser. "MemSum-DQA: Adapting An Efficient Long Document Extractive Summarizer for Document Question Answering." arXiv preprint arXiv:2310.06436 (2023). https://arxiv.org/abs/2310.06436
Abstract
We introduce MemSum-DQA, an efficient system for document question answering (DQA) that leverages MemSum, a long document extractive summarizer. By prefixing each text block in the parsed document with the provided question and question type, MemSum-DQA selectively extracts text blocks as answers from documents. On full-document answering tasks, this approach yields a 9% improvement in exact match accuracy over prior state-of-the-art baselines. Notably, MemSum-DQA excels in addressing questions related to child-relationship understanding, underscoring the potential of extractive summarization techniques for DQA tasks.