MemSum-DQA: Adapting An Efficient Long Document Extractive Summarizer for Document Question Answering

Nianlong Gu, Yingqiang Gao, Richard H. R. Hahnloser
10/10/2023
cs.CL

No official code found. Best implementation: nianlonggu/MemSum-DQA (10 stars, Apache-2.0, last commit 2023-10-14).

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.

View on arXivView PDF

Code Implementations(1)

nianlonggu/MemSum-DQACommunity100%
Apache-2.0

Adapting an Efficient Long Document Extractive Summarizer for Document Question Answering

101Jupyter NotebookOct 6, 20233 years agoApache-2.0

Similar Papers

Concise Answers to Complex Questions: Summarization of Long-form Answers

84% match1 repo72023

BERAG: Bayesian Ensemble Retrieval-Augmented Generation for Knowledge-based Visual Question Answering

82% match2026

Generator-Retriever-Generator Approach for Open-Domain Question Answering

81% match1 repo102023

Summarization-Based Document IDs for Generative Retrieval with Language Models

80% match1 repo2023

DTCRS: Dynamic Tree Construction for Recursive Summarization

80% match2026

AttenWalker: Unsupervised Long-Document Question Answering via Attention-based Graph Walking

80% match1 repo62023

Cite this paper

@article{gu2023memsumdqa,
  title  = {MemSum-DQA: Adapting An Efficient Long Document Extractive Summarizer for Document Question Answering},
  author = {Nianlong Gu and Yingqiang Gao and Richard H. R. Hahnloser},
  year   = {2023},
  eprint = {2310.06436},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2310.06436v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2310.06436.svg)](https://codeofpaper.com/paper/2310.06436)

Discussion