Extract with Order for Coherent Multi-Document Summarization

TL;DR

This paper presents an extractive multi-document summarization method that improves informativeness and coherence by combining rank-based sentence selection with a coherence model, validated on DUC 2004 datasets.

Contribution

It introduces a novel approach integrating continuous vector representations, key-phrases, and a coherence model for enhanced multi-document summarization.

Findings

Significant improvements in ROUGE scores over state-of-the-art methods

Enhanced summary coherence and readability

Effective use of vector representations and key-phrases

Abstract

In this work, we aim at developing an extractive summarizer in the multi-document setting. We implement a rank based sentence selection using continuous vector representations along with key-phrases. Furthermore, we propose a model to tackle summary coherence for increasing readability. We conduct experiments on the Document Understanding Conference (DUC) 2004 datasets using ROUGE toolkit. Our experiments demonstrate that the methods bring significant improvements over the state of the art methods in terms of informativity and coherence.