# A Supervised Approach to Extractive Summarisation of Scientific Papers

## TL;DR

This paper introduces a new dataset for scientific paper summarisation, leveraging author summaries, and develops models that effectively encode sentence context, significantly outperforming baseline methods.

## Contribution

The paper presents a novel dataset for scientific paper summarisation and proposes models that combine neural encoding with traditional features, improving performance.

## Key findings

- Models encoding sentence context outperform baselines.
- The new dataset enables research in scientific summarisation.
- Neural approaches benefit from local and global context encoding.

## Abstract

Automatic summarisation is a popular approach to reduce a document to its main arguments. Recent research in the area has focused on neural approaches to summarisation, which can be very data-hungry. However, few large datasets exist and none for the traditionally popular domain of scientific publications, which opens up challenging research avenues centered on encoding large, complex documents. In this paper, we introduce a new dataset for summarisation of computer science publications by exploiting a large resource of author provided summaries and show straightforward ways of extending it further. We develop models on the dataset making use of both neural sentence encoding and traditionally used summarisation features and show that models which encode sentences as well as their local and global context perform best, significantly outperforming well-established baseline methods.