The preprocessing of ground-truth labels for extractive summarization is based on this repo.
- WikiHowQA: Only adopt the positive QA samples as the question-driven summarization data
- PubMedQA: The dataset split for the question-driven summarization can be downloaded via the following url: https://drive.google.com/file/d/1K3sfU3u2pNIu2xd22LiPEyiiQHTZMMnc/view?usp=sharing
If the code is used in your research, please star this repo and cite our paper as follows:
@inproceedings{DBLP:conf/sigir/DengZL0LS20,
author = {Yang Deng and
Wenxuan Zhang and
Yaliang Li and
Min Yang and
Wai Lam and
Ying Shen},
title = {Bridging Hierarchical and Sequential Context Modeling for Question-driven
Extractive Answer Summarization},
booktitle = {Proceedings of the 43rd International {ACM} {SIGIR} conference on
research and development in Information Retrieval, {SIGIR} 2020, Virtual
Event, China, July 25-30, 2020},
pages = {1693--1696},
publisher = {{ACM}},
year = {2020},
}