Query-Based Abstractive Summarization Using Neural Networks

TL;DR

This paper introduces a neural network model for query-based abstractive summarization, enabling targeted summaries of documents in response to specific queries, trained and evaluated on adapted news dataset.

Contribution

It adapts neural network models for query-based summarization, demonstrating their effectiveness in generating targeted summaries using a pointer-generator approach.

Findings

Neural network models can produce query-specific summaries

The model achieves similarity to reference summaries

Query-based summarization improves relevance of generated summaries

Abstract

In this paper, we present a model for generating summaries of text documents with respect to a query. This is known as query-based summarization. We adapt an existing dataset of news article summaries for the task and train a pointer-generator model using this dataset. The generated summaries are evaluated by measuring similarity to reference summaries. Our results show that a neural network summarization model, similar to existing neural network models for abstractive summarization, can be constructed to make use of queries to produce targeted summaries.