Affiliation of Author(s):Beijing University of Posts and Telecommunications

Teaching and Research Group:Center for Intelligence Science and Technology

Journal:roceedings of the 23rd Conference on Computational Natural Language Learning

Key Words:abstractive summarization

Abstract:Various Seq2Seq learning models designed for machine translation were applied for abstractive summarization task recently. Despite these models provide high ROUGE scores, they are limited to generate comprehensive summaries with a high level of abstraction due to its degenerated attention distribution. We introduce Diverse Convolutional Seq2Seq Model(DivCNN Seq2Seq) using Determinantal Point Processes methods(Micro DPPs and Macro DPPs) to produce attention distribution considering both quality and diversity.

Indexed by:Essay collection

First Author:Lei Li

All the Authors:Marina Litvak,Natalia Vanetik,Zuying Huang

Correspondence Author:Weil Liu

Discipline:Engineering

First-Level Discipline:Computer science and technology *

Document Type:C

Page Number:822–832

Number of Words:6000

Translation or Not:no

Date of Publication:2019-11-03