Numerical reasoning over texts, such as addition, subtraction, sorting and counting, is a challenging machine reading comprehension task, since it requires both natural language understanding and arithmetic computation. To address this challenge, we propose a heterogeneous graph representation for the context of the passage and question needed for such reasoning, and design a question directed graph attention network to drive multi-step numerical reasoning over this context graph. The code link is at: https://github.com/emnlp2020qdgat/QDGAT
@article{arxiv.2009.07448,
title = {Question Directed Graph Attention Network for Numerical Reasoning over Text},
author = {Kunlong Chen and Weidi Xu and Xingyi Cheng and Zou Xiaochuan and Yuyu Zhang and Le Song and Taifeng Wang and Yuan Qi and Wei Chu},
journal= {arXiv preprint arXiv:2009.07448},
year = {2023}
}