We introduce fastcoref, a python package for fast, accurate, and easy-to-use English coreference resolution. The package is pip-installable, and allows two modes: an accurate mode based on the LingMess architecture, providing state-of-the-art coreference accuracy, and a substantially faster model, F-coref, which is the focus of this work. \model{} allows to process 2.8K OntoNotes documents in 25 seconds on a V100 GPU (compared to 6 minutes for the LingMess model, and to 12 minutes of the popular AllenNLP coreference model) with only a modest drop in accuracy. The fast speed is achieved through a combination of distillation of a compact model from the LingMess model, and an efficient batching implementation using a technique we call leftover batching. https://github.com/shon-otmazgin/fastcoref
翻译:我们引入了快速、准确和易于使用的英语引用解析法的快速芯片包。 包件是可安装的, 允许两种模式: 一种基于 LingMess 结构的精确模式, 提供最新的共同参考精度, 和一个大大加快的模型F- Coref, 这是这项工作的重点 。\ model\\ 允许在 V100 GPU (LingMess 模型为6分钟,流行的 AllenNLP 共同参照模型为12分钟) 处理 2.8K OntoNotes 文件25 秒后, 精度略下降 。 快速的速度是通过精炼LingMess 模型的紧凑模型, 以及使用我们称之为剩余批量的技术进行高效分批实施。 https:// github. com/ shon- otmazgin/ fastcoref 组合而实现的快速速度。