A marker collaborating model for entity and relation extraction

Recognizing entities and extracting their relations are significant tasks for information extraction. Many recent works focus on leveraging entity markers for better span (pair) representations, which have achieved promising performance. Nevertheless, existing works have two shortcomings: (1) previous markers are randomly embedded into distributed representations, which ignores semantic information relevant to the targeted tokens that we want to highlight. (2) most works simply implant markers into a sentence, which is weak to encode the interrelation between multiple tokens. In this work, we propose a marker collaborating model for entity and relation extraction, which is composed of two modules. The first module is semantic marker module, where two kinds of semantic markers are proposed to leverage semantic features from the PLMs. The second module is collaborative marker attention module, where a marker-oriented directional attention is proposed to model the interaction between multiple tokens. Our model yields SOTA performance on three standard benchmarks (ACE04, ACE05 and SciERC).

Automated summary

Recognizing entities and extracting their relations are important tasks in information extraction. Recent works have focused on leveraging entity markers for better span representations and have achieved promising performance. However, existing works have two shortcomings: (1) previous markers are randomly embedded in distributed representations, ignoring semantic information relevant to the targeted tokens; (2) most works simply implant markers into a sentence, lacking the ability to encode the interrelation between multiple tokens. This work proposes a marker collaborating model for entity and relation extraction, consisting of two modules, and achieves state-of-the-art performance on three standard benchmarks (ACE04, ACE05, and SciERC).