Recursive Top-Down Production for Sentence Generation with Latent Trees
Abstract
We model the recursive production property of context-free grammars for natural and synthetic languages. To this end, we present a dynamic programming algorithm that marginalises over latent binary tree structures with leaves, allowing us to compute the likelihood of a sequence of tokens under a latent tree model, which we maximise to train a recursive neural function. We demonstrate performance on two synthetic tasks: SCAN (Lake and Baroni, 2017), where it outperforms previous models on the LENGTH split, and English question formation (McCoy et al., 2020), where it performs comparably to decoders with the ground-truth tree structure. We also present experimental results on German-English translation on the Multi30k dataset (Elliott et al., 2016), and qualitatively analyse the induced tree structures our model learns for the SCAN tasks and the German-English translation task.
Cite
@article{arxiv.2010.04704,
title = {Recursive Top-Down Production for Sentence Generation with Latent Trees},
author = {Shawn Tan and Yikang Shen and Timothy J. O'Donnell and Alessandro Sordoni and Aaron Courville},
journal= {arXiv preprint arXiv:2010.04704},
year = {2020}
}