When to Use What: An In-Depth Comparative Empirical Analysis of OpenIE Systems for Downstream Applications

TL;DR

This paper provides an empirical comparison of neural OpenIE models, training sets, and benchmarks to guide users in selecting suitable systems for various NLP downstream tasks, emphasizing the importance of model assumptions.

Contribution

It offers an application-focused empirical survey that analyzes how different models and datasets impact OpenIE performance, aiding in informed system selection.

Findings

Model assumptions significantly affect performance

Training set differences influence model effectiveness

Recommendations improve downstream Complex QA results

Abstract

Open Information Extraction (OpenIE) has been used in the pipelines of various NLP tasks. Unfortunately, there is no clear consensus on which models to use in which tasks. Muddying things further is the lack of comparisons that take differing training sets into account. In this paper, we present an application-focused empirical survey of neural OpenIE models, training sets, and benchmarks in an effort to help users choose the most suitable OpenIE systems for their applications. We find that the different assumptions made by different models and datasets have a statistically significant effect on performance, making it important to choose the most appropriate model for one's applications. We demonstrate the applicability of our recommendations on a downstream Complex QA application.