Enriched Attention for Robust Relation Extraction

TL;DR

This paper introduces an enriched attention mechanism for neural relation extraction that improves robustness and scalability to long sentences with multiple entities by incorporating relation argument features and dependency paths.

Contribution

It proposes a novel enriched attention method that enhances focus on relevant sentence parts, outperforming prior models on benchmark datasets.

Findings

Model outperforms prior work on benchmarks.

Enriched attention improves focus on relevant sentence parts.

Scales effectively to long sentences with many entities.

Abstract

The performance of relation extraction models has increased considerably with the rise of neural networks. However, a key issue of neural relation extraction is robustness: the models do not scale well to long sentences with multiple entities and relations. In this work, we address this problem with an enriched attention mechanism. Attention allows the model to focus on parts of the input sentence that are relevant to relation extraction. We propose to enrich the attention function with features modeling knowledge about the relation arguments and the shortest dependency path between them. Thus, for different relation arguments, the model can pay attention to different parts of the sentence. Our model outperforms prior work using comparable setups on two popular benchmarks, and our analysis confirms that it indeed scales to long sentences with many entities.