Event-centric structured prediction involves predicting structured outputs of events. In most NLP cases, event structures are complex with manifold dependency, and it is challenging to effectively represent these complicated structured events. To address these issues, we propose Structured Prediction with Energy-based Event-Centric Hyperspheres (SPEECH). SPEECH models complex dependency among event structured components with energy-based modeling, and represents event classes with simple but effective hyperspheres. Experiments on two unified-annotated event datasets indicate that SPEECH is predominant in event detection and event-relation extraction tasks.

翻译：事件为中心的结构化预测涉及对事件的结构化输出进行预测。在大多数自然语言处理案例中，事件结构复杂且具有多层面依赖性，有效表征这些复杂的结构化事件极具挑战性。为解决这一问题，我们提出基于能量的以事件为中心的超球面结构化预测（SPEECH）。SPEECH利用基于能量的建模对事件结构化组件间的复杂依赖关系进行建模，并通过简单有效的超球面表示事件类别。在两个统一标注的事件数据集上的实验表明，SPEECH在事件检测和事件关系抽取任务中表现优异。