Enriched Attention for Robust Relation Extraction
This addresses robustness issues in relation extraction for NLP applications, but it is incremental as it builds on existing attention mechanisms.
The paper tackled the problem of neural relation extraction models lacking robustness on long sentences with multiple entities and relations by proposing an enriched attention mechanism incorporating features like relation arguments and dependency paths, resulting in outperforming prior work on two benchmarks and scaling effectively to long sentences.
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.