Non-neural Models Matter: A Re-evaluation of Neural Referring Expression Generation Systems

TL;DR

This paper demonstrates that well-designed non-neural models can outperform neural approaches in referring expression generation, emphasizing the importance of traditional methods alongside neural models.

Contribution

The study re-evaluates the performance of non-neural versus neural models in referring expression generation, showing non-neural models can be competitive or superior.

Findings

Rule-based systems perform on par or better than neural models on datasets.

A well-designed machine learning system with linguistic features outperforms neural models on WSJ.

Human evaluations support the effectiveness of non-neural approaches.

Abstract

In recent years, neural models have often outperformed rule-based and classic Machine Learning approaches in NLG. These classic approaches are now often disregarded, for example when new neural models are evaluated. We argue that they should not be overlooked, since, for some tasks, well-designed non-neural approaches achieve better performance than neural ones. In this paper, the task of generating referring expressions in linguistic context is used as an example. We examined two very different English datasets (WEBNLG and WSJ), and evaluated each algorithm using both automatic and human evaluations. Overall, the results of these evaluations suggest that rule-based systems with simple rule sets achieve on-par or better performance on both datasets compared to state-of-the-art neural REG systems. In the case of the more realistic dataset, WSJ, a machine learning-based system with well-designed linguistic features performed best. We hope that our work can encourage researchers to consider non-neural models in future.