Dynamic Multi-View Fusion Mechanism for Chinese Relation Extraction

Abstract Recently, many studies incorporate external knowledge into character-level feature based models to improve the performance of Chinese relation extraction. However, these methods tend ignore internal information character and cannot filter out noisy knowledge. To address issues, we propose a mixture-of-view-experts framework (MoVE) dynamically learn multi-view features for With both characters, our can better capture semantic characters. demonstrate effectiveness proposed framework, conduct extensive experiments on three real-world datasets in distinct domains. Experimental results show consistent significant superiority robustness framework. Our code dataset will be released at: https://gitee.com/tmg-nudt/multi-view-of-expert-for-chinese-relation-extraction