GDPNet：精炼用于关系抽取的潜在多视图图
GDPNet: Refining Latent Multi-View Graph for Relation Extraction
AI总结：
本文提出GDPNet，通过潜在多视图图、GGG建边与DTWPool精炼筛选关系指示词，并融合BERT表示，在对话级关系抽取上达到最佳性能。
摘要：
关系抽取（Relation Extraction，RE）旨在预测一段文本（例如一个句子或一段对话）中提及的两个实体之间的关系类型。当给定文本较长时，识别用于关系预测的指示性词语具有挑战性。RE任务的最新进展来自基于BERT的序列建模，以及对序列中 token 之间关系的基于图的建模。本文提出构建一个潜在多视图图，以捕捉 token 之间的各种可能关系。随后精炼该图，以选择对关系预测重要的词语。最后，将精炼后图的表示与基于BERT的序列表示拼接，用于关系抽取。具体而言，在所提出的GDPNet（Gaussian Dynamic Time Warping Pooling Net，高斯动态时间规整池化网络）中，利用Gaussian Graph Generator（GGG，高斯图生成器）生成多视图图的边，然后通过Dynamic Time Warping Pooling（DTWPool，动态时间规整池化）精炼该图。在DialogRE和TACRED上的实验表明，GDPNet在对话级RE上取得最佳性能，并在句子级RE上取得与最先进方法相当的性能。
英文摘要：
Relation Extraction (RE) is to predict the relation type of two entities that are mentioned in a piece of text, e.g., a sentence or a dialogue. When the given text is long, it is challenging to identify indicative words for the relation prediction. Recent advances on RE task are from BERT-based sequence modeling and graph-based modeling of relationships among the tokens in the sequence. In this paper, we propose to construct a latent multi-view graph to capture various possible relationships among tokens. We then refine this graph to select important words for relation prediction. Finally, the representation of the refined graph and the BERT-based sequence representation are concatenated for relation extraction. Specifically, in our proposed GDPNet (Gaussian Dynamic Time Warping Pooling Net), we utilize Gaussian Graph Generator (GGG) to generate edges of the multi-view graph. The graph is then refined by Dynamic Time Warping Pooling (DTWPool). On DialogRE and TACRED, we show that GDPNet achieves the best performance on dialogue-level RE, and comparable performance with the state-of-the-arts on sentence-level RE.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
