OpenIE6: Iterative Grid Labeling and Coordination Analysis for Open Information Extraction

Abstract

A recent state-of-the-art neural open information extraction (OpenIE) system generates extractions iteratively, requiring repeated encoding of partial outputs. This comes at a significant computational cost. On the other hand, sequence labeling approaches for OpenIE are much faster, but worse in extraction quality. In this paper, we bridge this trade-off by presenting an iterative labeling-based system that establishes a new state of the art for OpenIE, while extracting 10× faster. This is achieved through a novel Iterative Grid Labeling (IGL) architecture, which treats OpenIE as a 2-D grid labeling task. We improve its performance further by applying coverage (soft) constraints on the grid at training time.Moreover, on observing that the best OpenIE systems falter at handling coordination structures, our OpenIE system also incorporates a new coordination analyzer built with the same IGL architecture. This IGL based coordination analyzer helps our OpenIE system handle complicated coordination structures, while also establishing a new state of the art on the task of coordination analysis, with a 12.3 pts improvement in F1 over previous analyzers. Our OpenIE system, OpenIE6 1 , beats the previous systems by as much as 4 pts in F1, while being much faster.

Biological Insights Knowledge Graph: an integrated knowledge graph to support drug development

“…While the data imported from various structured sources provides the most reliable part of the graph, it is enriched using large-scale relation extraction from free-text sources such as scientific literature from PubMed 1 . The NLP aspects of BIKG is based on a series of pipelines, ranging from simple entity co-occurrence and traditional rule based dependency parsing, to state-of-the-art relationship classification with RBERT [49] and open information extraction with OpenIE6 [26] neural information extraction system. In terms of quantity, this NLP-extracted data constitutes the largest component of the graph, providing around 80% of graph edges.…”

LILLIE: Information extraction and database integration using linguistics and learning-based algorithms

“…To evaluate our system, we first measure the performance of our triple extractor against two state-of-the-art systems, OpenIE6 [55] and IMoJIE [56], on two standard benchmark data sets. Next, we use the PubMed abstracts dataset to demonstrate the qualitative advantages of our enhancements, in comparison to these systems and to show that our approach generalizes well for a diverse set of datasets.…”

Zero-Shot Information Extraction as a Unified Text-to-Triple Translation

“…Many open information extraction (OIE) systems, e.g., Stanford OpenIE (Angeli et al, 2015), OLLIE (Schmitz et al, 2012), Reverb (Fader et al, 2011), and their descendant Open IE4 leverage carefully-designed linguistic patterns (e.g., based on dependencies and POS tags) to extract triples from textual corpora without using additional training sets. Recently, supervised OIE systems (Stanovsky et al, 2018;Ro et al, 2020;Kolluru et al, 2020) formulate the OIE as a sequence generation problem using neural networks trained on additional training sets. Similar to our work, Wang et al (2020) use the parameters of LMs to extract triples, with the main difference that DEEPEX not only improves the recall of the beam search, but also uses a pre-trained ranking model to enhance the zero-shot capability.…”