F-coref:快速，准确和易于使用的共同参考分辨率

Q3 Environmental Science

AACL Bioflux Pub Date : 2022-09-09 DOI:10.48550/arXiv.2209.04280

Shon Otmazgin, Arie Cattan, Yoav Goldberg

{"title":"F-coref:快速，准确和易于使用的共同参考分辨率","authors":"Shon Otmazgin, Arie Cattan, Yoav Goldberg","doi":"10.48550/arXiv.2209.04280","DOIUrl":null,"url":null,"abstract":"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. F-coref 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","PeriodicalId":39298,"journal":{"name":"AACL Bioflux","volume":"47 1","pages":"48-56"},"PeriodicalIF":0.0000,"publicationDate":"2022-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"7","resultStr":"{\"title\":\"F-coref: Fast, Accurate and Easy to Use Coreference Resolution\",\"authors\":\"Shon Otmazgin, Arie Cattan, Yoav Goldberg\",\"doi\":\"10.48550/arXiv.2209.04280\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"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. F-coref 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\",\"PeriodicalId\":39298,\"journal\":{\"name\":\"AACL Bioflux\",\"volume\":\"47 1\",\"pages\":\"48-56\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2022-09-09\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"7\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"AACL Bioflux\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.48550/arXiv.2209.04280\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"Environmental Science\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"AACL Bioflux","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.48550/arXiv.2209.04280","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"Environmental Science","Score":null,"Total":0}

引用次数: 7

摘要

我们介绍fastcoref，这是一个python包，用于快速、准确和易于使用的英语共同参考解析。该包是可pip安装的，并允许两种模式:基于LingMess架构的精确模式，提供最先进的共参考精度，以及更快的模型F-coref，这是本工作的重点。F-coref允许在V100 GPU上在25秒内处理2.8K OntoNotes文档(相比之下，LingMess模型需要6分钟，流行的AllenNLP共同参考模型需要12分钟)，精度只有轻微下降。快速的速度是通过结合LingMess模型的精简模型的蒸馏，以及使用我们称之为剩余批处理技术的高效批处理实现来实现的。https://github.com/shon-otmazgin/fastcoref

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

F-coref: Fast, Accurate and Easy to Use Coreference Resolution

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. F-coref 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

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

AACL Bioflux Environmental Science-Management, Monitoring, Policy and Law

CiteScore

1.40

自引率

0.00%

发文量