A Noisy-Channel Model for Document Compression

TL;DR

This paper introduces a hierarchical noisy-channel model for document compression that leverages syntactic and discourse structures to produce coherent, grammatical summaries, outperforming simpler baseline methods.

Contribution

It presents a novel hierarchical model that incorporates discourse and syntactic structures for improved document compression.

Findings

The system outperforms baseline and sentence-based compression methods.

Discourse knowledge significantly improves document summarization quality.

Hierarchical modeling captures essential information for coherent compression.

Abstract

We present a document compression system that uses a hierarchical noisy-channel model of text production. Our compression system first automatically derives the syntactic structure of each sentence and the overall discourse structure of the text given as input. The system then uses a statistical hierarchical model of text production in order to drop non-important syntactic and discourse constituents so as to generate coherent, grammatical document compressions of arbitrary length. The system outperforms both a baseline and a sentence-based compression system that operates by simplifying sequentially all sentences in a text. Our results support the claim that discourse knowledge plays an important role in document summarization.