We introduce the first end-to-end coreference resolution model and show that
it significantly outperforms all previous work without using a syntactic parser
or hand-engineered mention detector. The key idea is to directly consider all
spans in a document as potential mentions and learn distributions over possible
antecedents for each. The model computes span embeddings that combine
context-dependent boundary representations with a head-finding attention
mechanism. It is trained to maximize the marginal likelihood of gold antecedent
spans from coreference clusters and is factored to enable aggressive pruning of
potential mentions. Experiments demonstrate state-of-the-art performance, with
a gain of 1.5 F1 on the OntoNotes benchmark and by 3.1 F1 using a 5-model
ensemble, despite the fact that this is the first approach to be successfully
trained with no external resources.

本文介绍了第一个端到端的共指消解模型，并展示了它在不使用句法分析器或手工制作的提及检测器的情况下，明显优于所有以前的工作。模型的关键思想是直接将文档中的所有跨度作为可能的提及，并学习每个可能的先行词的分布。模型计算跨度嵌入，将依赖于上下文的边界表示与定位注意机制相结合。它被训练成最大化核对聚类中的黄金先行跨度的边际可能性，并分解以便于对可能存在的提及进行积极修剪。实验表明，尽管这是第一个成功训练不使用外部资源的方法，它仍可以在 OntoNotes 基准测试上获得 1.5 F1 分数的增益，并使用 5 个模型集合获得 3.1 F1 分数的提升。