IMoJIE: Iterative Memory-Based Joint Open Information Extraction

TL;DR

IMoJIE is a novel neural model for open information extraction that iteratively produces diverse extractions conditioned on previous outputs, significantly improving over prior neural approaches.

Contribution

It introduces an iterative, memory-based extension to CopyAttention, enabling variable and diverse extractions, and establishes new state-of-the-art results.

Findings

IMoJIE outperforms CopyAttention by about 18 F1 points.

IMoJIE surpasses a BERT-based baseline by 2 F1 points.

Training on filtered, bootstrapped data enhances extraction quality.

Abstract

While traditional systems for Open Information Extraction were statistical and rule-based, recently neural models have been introduced for the task. Our work builds upon CopyAttention, a sequence generation OpenIE model (Cui et. al., 2018). Our analysis reveals that CopyAttention produces a constant number of extractions per sentence, and its extracted tuples often express redundant information. We present IMoJIE, an extension to CopyAttention, which produces the next extraction conditioned on all previously extracted tuples. This approach overcomes both shortcomings of CopyAttention, resulting in a variable number of diverse extractions per sentence. We train IMoJIE on training data bootstrapped from extractions of several non-neural systems, which have been automatically filtered to reduce redundancy and noise. IMoJIE outperforms CopyAttention by about 18 F1 pts, and a BERT-based strong baseline by 2 F1 pts, establishing a new state of the art for the task.