Relational Gating for "What If" Reasoning
Abstract
This paper addresses the challenge of learning to do procedural reasoning over text to answer "What if..." questions. We propose a novel relational gating network that learns to filter the key entities and relationships and learns contextual and cross representations of both procedure and question for finding the answer. Our relational gating network contains an entity gating module, relation gating module, and contextual interaction module. These modules help in solving the "What if..." reasoning problem. We show that modeling pairwise relationships helps to capture higher-order relations and find the line of reasoning for causes and effects in the procedural descriptions. Our proposed approach achieves the state-of-the-art results on the WIQA dataset.
Cite
Text
Zheng and Kordjamshidi. "Relational Gating for "What If" Reasoning." International Joint Conference on Artificial Intelligence, 2021. doi:10.24963/IJCAI.2021/553Markdown
[Zheng and Kordjamshidi. "Relational Gating for "What If" Reasoning." International Joint Conference on Artificial Intelligence, 2021.](https://mlanthology.org/ijcai/2021/zheng2021ijcai-relational/) doi:10.24963/IJCAI.2021/553BibTeX
@inproceedings{zheng2021ijcai-relational,
title = {{Relational Gating for "What If" Reasoning}},
author = {Zheng, Chen and Kordjamshidi, Parisa},
booktitle = {International Joint Conference on Artificial Intelligence},
year = {2021},
pages = {4015-4022},
doi = {10.24963/IJCAI.2021/553},
url = {https://mlanthology.org/ijcai/2021/zheng2021ijcai-relational/}
}