Generating Summaries with Topic Templates and Structured Convolutional Decoders

06/11/2019
by   Laura Perez-Beltrachini, et al.
0

Existing neural generation approaches create multi-sentence text as a single sequence. In this paper we propose a structured convolutional decoder that is guided by the content structure of target summaries. We compare our model with existing sequential decoders on three data sets representing different domains. Automatic and human evaluation demonstrate that our summaries have better content coverage.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
03/05/2018

ROUGE 2.0: Updated and Improved Measures for Evaluation of Summarization Tasks

Evaluation of summarization tasks is extremely crucial to determining th...
research
12/01/2015

Inferring Interpersonal Relations in Narrative Summaries

Characterizing relationships between people is fundamental for the under...
research
05/10/2018

Regularizing Output Distribution of Abstractive Chinese Social Media Text Summarization for Improved Semantic Consistency

Abstractive text summarization is a highly difficult problem, and the se...
research
08/01/2018

From Thumbnails to Summaries - A single Deep Neural Network to Rule Them All

Video summaries come in many forms, from traditional single-image thumbn...
research
09/02/2019

Collecting and Structuring Information in the Information Collage

Knowledge workers, such as scientists, journalists, or consultants, adap...
research
02/23/2020

Path Outlines: Browsing Path-Based Summaries of Linked Open Datasets

Linked Data (LD) are structured sources of information, such as DBpedia ...
research
05/06/2019

Toward Human-Like Summaries Generated from Heterogeneous Software Artefacts

Automatic text summarisation has drawn considerable interest in the fiel...

Please sign up or login with your details

Forgot password? Click here to reset