Latent Tree Learning with Differentiable Parsers: Shift-Reduce Parsing and Chart Parsing
Jean Pierre Maillard, Stephen Charles Clark · 2018
Latent tree learning models represent sentences by composing their words according to an induced parse tree, all based on a downstream task.These models often outperform baselines which use (externally provided) syntax trees to drive the composition order.This work contributes (a) a new latent tree learning model based on shift-reduce parsing, with competitive downstream performance and non-trivial induced trees, and (b) an analysis of the trees learned by our shift-reduce model and by a chart-based model.