英国爱丁堡大学刘洋博士学术报告

发布日期:2019-10-08 浏览次数: [字体: ]

报告题目:Neural Document Modeling and Summarization

报告人:      刘洋  博士   英国爱丁堡大学

报告时间:2019. 10. 9(周三)下午3:30 

报告地点:明理楼402报告厅 

报告内容简介:

With the success of neural models in sentence-level and paragraph-level NLP tasks, increasing attention has been drawn to document-level tasks like document classification and summarization. Many methods have been proposed to better encode a document. But due to the long-length and complex discourse structure, document modeling is still a challenging problem. In this talk, I will present my PhD work in neural document modeling and summarization, from imposing latent structured bias to applying pretrained models like BERT. I will also discuss the multi-document summarization task and a new hierarchical approach with the capability to encode multiple documents.

刘洋简介: 

Yang Liu is a PhD student at the University of Edinburgh, supervised by Professor Mirella Lapata. His research interests are Document Modelling and Text Summarization.  He was awarded the Google PhD Fellowship at 2018. Prior to his PhD study, Yang got his master degree from Peking University and bachelor degree from Tianjin University in China.