When Are Tree Structures Necessary for Deep Learning of Representations?

TL;DR

This paper evaluates when recursive neural models outperform recurrent models by benchmarking on four NLP tasks, revealing recursive models excel in long-distance relation tasks and proposing a method to enhance recurrent models.

Contribution

It provides a rigorous comparison between recursive and recurrent models across multiple tasks and introduces a technique to improve recurrent models' performance on long sequences.

Findings

Recursive models outperform on long-distance relation tasks.

Breaking sentences into clauses improves recurrent models.

Recursive models are better for tasks requiring long-distance associations.

Abstract

Recursive neural models, which use syntactic parse trees to recursively generate representations bottom-up, are a popular architecture. But there have not been rigorous evaluations showing for exactly which tasks this syntax-based method is appropriate. In this paper we benchmark {\bf recursive} neural models against sequential {\bf recurrent} neural models (simple recurrent and LSTM models), enforcing apples-to-apples comparison as much as possible. We investigate 4 tasks: (1) sentiment classification at the sentence level and phrase level; (2) matching questions to answer-phrases; (3) discourse parsing; (4) semantic relation extraction (e.g., {\em component-whole} between nouns). Our goal is to understand better when, and why, recursive models can outperform simpler models. We find that recursive models help mainly on tasks (like semantic relation extraction) that require associating headwords across a long distance, particularly on very long sequences. We then introduce a method for allowing recurrent models to achieve similar performance: breaking long sentences into clause-like units at punctuation and processing them separately before combining. Our results thus help understand the limitations of both classes of models, and suggest directions for improving recurrent models.