论文标题
在文档级联合实体和关系提取中建模任务交互
Modeling Task Interactions in Document-Level Joint Entity and Relation Extraction
论文作者
论文摘要
我们在端到端设置中针对文档级的关系提取，在该设置中，模型需要共同执行提取，核心分辨率（coref）和关系提取（RE），并以实体中心的方式进行评估。特别是，我们解决了CoreF和RE之间的双向交互，这不是先前工作的重点，并建议引入明确的互动，即图形兼容性（GC），该互动是专门设计用于利用任务特征的，桥接任务的决策，以实现直接任务干扰。我们的实验是在Docred和dwie上进行的。除GC外，我们还实施并比较了以前工作中通常采用的不同多任务设置，包括管道，共享编码器，图形传播，以检查不同相互作用的有效性。结果表明，GC的表现最佳，高达2.3/5.1 F1比基线提高。
We target on the document-level relation extraction in an end-to-end setting, where the model needs to jointly perform mention extraction, coreference resolution (COREF) and relation extraction (RE) at once, and gets evaluated in an entity-centric way. Especially, we address the two-way interaction between COREF and RE that has not been the focus by previous work, and propose to introduce explicit interaction namely Graph Compatibility (GC) that is specifically designed to leverage task characteristics, bridging decisions of two tasks for direct task interference. Our experiments are conducted on DocRED and DWIE; in addition to GC, we implement and compare different multi-task settings commonly adopted in previous work, including pipeline, shared encoders, graph propagation, to examine the effectiveness of different interactions. The result shows that GC achieves the best performance by up to 2.3/5.1 F1 improvement over the baseline.
