Few-Shot Semantic Parsing for New Predicates
Published in EACL, 2021
Recommended citation: Li, Z., Qu, L., Huang, S., & Haffari, G. Few-shot Learning for Semantic Parsing. 16th conference of the European Chapter of the Association for Computational Linguistics https://arxiv.org/abs/2101.10708
In this work, we investigate the problems of semantic parsing in a few-shot learning setting. In this setting, we are provided with utterance-logical form pairs per new predicate. The state-of-the-art neural semantic parsers achieve less than 25% accuracy on benchmark datasets when k= 1. To tackle this problem, we proposed to i) apply a designated meta-learning method to train the model; ii) regularize attention scores with alignment statistics; iii) apply a smoothing technique in pre-training. As a result, our method consistently outperforms all the baselines in both one and two-shot settings.
Citation:
@article{li2021few,
title={Few-Shot Semantic Parsing for New Predicates},
author={Li, Zhuang and Qu, Lizhen and Huang, Shuo and Haffari, Gholamreza},
journal={arXiv preprint arXiv:2101.10708},
year={2021}
}