arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2107.01700cs.CL
收藏

端到端神经共指消解的再探讨:一个简单而有效的基线

End-to-end Neural Coreference Resolution Revisited: A Simple yet Effective Baseline

University of Illinois at Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校);Adobe Research(奥多比研究院);Roku Inc.(Roku公司)

Tuan Manh Lai, Trung Bui, Doo Soon Kim

摘要:

自从第一个端到端神经共指消解模型被引入以来,已经提出了对该模型的许多扩展,从使用高阶推理到使用强化学习直接优化评估指标。尽管提高了共指消解性能的一个很大的利润,这些扩展增加了大量的额外的复杂性,原来的模型。基于这一观察结果以及预训练Transformer语言模型的最新进展,我们提出了一个简单而有效的共指消解基线。尽管我们的模型是原始神经共指消解模型的简化版本,但它取得了令人印象深刻的性能,在公共英语OntoNotes基准测试中优于所有最近的扩展作品。我们的工作提供了证据,仔细证明现有的或新提出的模型的复杂性的必要性,作为引入一个概念或实际的简化现有的模型仍然可以产生竞争力的结果。

英文摘要:

Since the first end-to-end neural coreference resolution model was introduced, many extensions to the model have been proposed, ranging from using higher-order inference to directly optimizing evaluation metrics using reinforcement learning. Despite improving the coreference resolution performance by a large margin, these extensions add substantial extra complexity to the original model. Motivated by this observation and the recent advances in pre-trained Transformer language models, we propose a simple yet effective baseline for coreference resolution. Even though our model is a simplified version of the original neural coreference resolution model, it achieves impressive performance, outperforming all recent extended works on the public English OntoNotes benchmark. Our work provides evidence for the necessity of carefully justifying the complexity of existing or newly proposed models, as introducing a conceptual or practical simplification to an existing model can still yield competitive results.

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。

↑