Rule Augmented Unsupervised Constituency Parsing
Abstract
Recently, unsupervised parsing of syntactic trees has gained considerable attention. A prototypical approach to such unsupervised parsing employs reinforcement learning and auto-encoders. However, no mechanism ensures that the learnt model leverages the well-understood language grammar. We propose an approach that utilizes very generic linguistic knowledge of the language present in the form of syntactic rules, thus inducing better syntactic structures. We introduce a novel formulation that takes advantage of the syntactic grammar rules and is independent of the base system. We achieve new state-of-the-art results on two benchmarks datasets, MNLI and WSJ. The source code of the paper is available at https://github.com/anshuln/Diora_with_rules.
Cite
@article{arxiv.2105.10193,
title = {Rule Augmented Unsupervised Constituency Parsing},
author = {Atul Sahay and Anshul Nasery and Ayush Maheshwari and Ganesh Ramakrishnan and Rishabh Iyer},
journal= {arXiv preprint arXiv:2105.10193},
year = {2021}
}
Comments
Accepted at Findings of ACL 2021. 10 Pages, 5 Tables, 2 Figures