The Role of Discourse Units in Near-Extractive Summarization

Although human-written summaries of documents tend to involve significant edits to the source text, most automated summa-rizers are extractive and select sentences verbatim. In this work we examine how elementary discourse units (EDUs) from Rhetorical Structure Theory can be used to extend extractive summarizers to produce a wider range of human-like summaries. Our analysis demonstrates that EDU segmentation is effective in preserving human-labeled summarization concepts within sentences and also aligns with near-extractive summaries constructed by news editors. Finally, we show that us-ing EDUs as units of content selection instead of sentences leads to stronger summarization performance in near-extractive scenarios, especially under tight budgets.

• Publication year

  2016

• Venue

  SIGDIAL Conference

• Publication date

  2016-09-01

• Fields of study

  Linguistics, Computer Science

• Identifiers
  DOI 10.18653/v1/W16-3617
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Toward Abstractive Summarization Using Semantic Representations

  2018cited by this paper
• PEAK: Pyramid Evaluation via Automated Knowledge Extraction

  2016cited by this paper
• Ranking with Recursive Neural Networks and Its Application to Multi-Document Summarization

  2015cited by this paper
• System Combination for Multi-document Summarization

  2015cited by this paper
• CODRA: A Novel Discriminative Framework for Rhetorical Analysis

  2015cited by this paper
• Abstractive Multi-document Summarization with Semantic Information Extraction

  2015cited by this paper
• Abstractive Multi-Document Summarization via Phrase Selection and Merging

  2015cited by this paper
• Single Document Summarization based on Nested Tree Structure

  2014influential reference
• Dependency-based Discourse Parser for Single-Document Summarization

  2014influential reference
• Two-pass Discourse Segmentation with Pairing and Global Features

  2014cited by this paper
• A Linear-Time Bottom-Up Discourse Parser with Constraints and Post-Editing

  2014cited by this paper
• Detecting Information-Dense Texts in Multiple News Domains

  2014cited by this paper
• Fast and Robust Compressive Summarization with Dual Decomposition and Multi-Task Learning

  2013cited by this paper
• Single-Document Summarization as a Tree Knapsack Problem

  2013cited by this paper
• Multiple Aspect Summarization Using Integer Linear Programming

  2012cited by this paper
• Structured Perceptron with Inexact Search

  2012cited by this paper
• A Class of Submodular Functions for Document Summarization

  2011cited by this paper
• Jointly Learning to Extract and Compress

  2011cited by this paper
• A Universal Part-of-Speech Tagset

  2011cited by this paper
• Discourse indicators for content selection in summarization

  2010cited by this paper
• HILDA: A Discourse Parser Using Support Vector Machine Classification

  2010cited by this paper
• A Scalable Global Model for Summarization

  2009cited by this paper
• The Penn Discourse TreeBank 2.0.

  2008influential reference
• A Framework for Identifying Textual Redundancy

  2008cited by this paper
• A Study of Global Inference Algorithms in Multi-document Summarization

  2007cited by this paper
• The Pyramid Method: Incorporating human content selection variation in summarization evaluation

  2007cited by this paper
• Modelling Compression with Discourse Constraints

  2007cited by this paper
• Automated Summarization Evaluation with Basic Elements.

  2006cited by this paper
• A Formal Model for Information Selection in Multi-Sentence Text Extraction

  2004cited by this paper
• ROUGE: A Package for Automatic Evaluation of Summaries

  2004cited by this paper
• Discriminative Training Methods for Hidden Markov Models: Theory and Experiments with Perceptron Algorithms

  2002influential reference
• Automatic summarization of English broadcast news speech

  2002cited by this paper
• A Noisy-Channel Model for Document Compression

  2002cited by this paper
• Building a Discourse-Tagged Corpus in the Framework of Rhetorical Structure Theory

  2001cited by this paper
• Cut and Paste Based Text Summarization

  2000cited by this paper
• Discourse Trees Are Good Indicators of Importance in Text

  1999influential reference
• The Use of MMR, Diversity-Based Reranking for Reordering Documents and Producing Summaries

  1998cited by this paper
• Rhetorical Structure Theory: Toward a functional theory of text organization

  1988cited by this paper
• Identification and Characterization of Newsworthy Verbs in World News

  year unknowncited by this paper

• Which course? Discourse! Teaching Discourse and Generation in the Era of LLMs

  2026cites this paper
• Synthetic Data Augmentation for Cross-domain Implicit Discourse Relation Recognition

  2025cites this paper
• A oposição nas relações de coerência Contrast e Concession

  2025cites this paper
• PreSumm: Predicting Summarization Performance Without Summarizing

  2025cites this paper
• CoMuMDR: Code-mixed Multi-modal Multi-domain corpus for Discourse paRsing in conversations

  2025cites this paper
• AspirinSum: an Aspect-based utility-preserved de-identification Summarization framework

  2024cites this paper
• RST-LoRA: A Discourse-Aware Low-Rank Adaptation for Long Document Abstractive Summarization

  2024cites this paper
• Abstractive Summarization based on Elemental Discourse Units using Different Attention Mechanisms

  2024cites this paper
• Extract-and-Abstract: Unifying Extractive and Abstractive Summarization within Single Encoder-Decoder Framework

  2024cites this paper
• ACUEval: Fine-grained Hallucination Evaluation and Correction for Abstractive Summarization

  2024cites this paper
• Generating EDU Extracts for Plan-Guided Summary Re-Ranking

  2023cites this paper
• Why Can’t Discourse Parsing Generalize? A Thorough Investigation of the Impact of Data Diversity

  2023cites this paper
• Explainable Artificial Intelligence for Highlighting and Searching in Patent Text 12-21

  2023cites this paper
• DBERT-ELVA: Discourse-Aware Extractive Text Summarization with Autoencoder

  2023cites this paper
• A survey of discourse parsing

  2022cites this paper
• Generating news image captions with semantic discourse extraction and contrastive style-coherent learning

  2022cites this paper
• Extractive Summarisation for German-language Data: A Text-level Approach with Discourse Features

  2022cites this paper
• EDU-level Extractive Summarization with Varying Summary Lengths

  2022cites this paper
• An Annotated Dataset and Automatic Approaches for Discourse Mode Identification in Low-resource Bengali Language

  2022cites this paper
• A topic modeled unsupervised approach to single document extractive text summarization

  2022cites this paper
• A Proposition-Level Clustering Approach for Multi-Document Summarization

  2021cites this paper
• Heterogeneous Graph Based Extractive Summarization Considering Discourse and Coreference Relations

  2021cites this paper
• Improving Unsupervised Extractive Summarization with Facet-Aware Modeling

  2021cites this paper
• Extractive Summarization Considering Discourse and Coreference Relations based on Heterogeneous Graph

  2021cites this paper
• Proposition-Level Clustering for Multi-Document Summarization

  2021cites this paper
• Manual Evaluations of Extractive Oracle Summaries

  2020cites this paper
• Text Segmentation by Cross Segment Attention

  2020cites this paper
• Exploring Content Selection in Summarization of Novel Chapters

  2020cites this paper
• An Annotated Dataset of Discourse Modes in Hindi Stories

  2020cites this paper
• Composing Elementary Discourse Units in Abstractive Summarization

  2020cites this paper
• Aspect Category Detection in Product Reviews using Contextual Representation

  2020cites this paper
• SuperPAL: Supervised Proposition ALignment for Multi-Document Summarization and Derivative Sub-Tasks

  2020cites this paper
• QADiscourse - Discourse Relations as QA Pairs: Representation, Crowdsourcing and Baselines

  2020cites this paper
• Compressive Summarization with Plausibility and Salience Modeling

  2020cites this paper
• Better Highlighting: Creating Sub-Sentence Summary Highlights

  2020cites this paper
• QBSUM: a Large-Scale Query-Based Document Summarization Dataset from Real-world Applications

  2020cites this paper
• Neural Methods for Sentiment Analysis and Text Summarization. (Méthodes neuronales pour l'analyse des sentiments et la synthèse des textes)

  2020influential citation
• Intrinsic Evaluation of Summarization Datasets

  2020cites this paper
• Diachronic Embeddings for People in the News

  2020cites this paper
• News Editorials: Towards Summarizing Long Argumentative Texts

  2020cites this paper
• Research on Discourse Parsing: from the Dependency View

  2020cites this paper
• Neural Extractive Text Summarization with Syntactic Compression

  2019cites this paper
• Discourse-Aware Neural Extractive Text Summarization

  2019cites this paper
• Exploiting Discourse-Level Segmentation for Extractive Summarization

  2019cites this paper
• Joint Parsing and Generation for Abstractive Summarization

  2019cites this paper
• Discourse-Aware Neural Extractive Model for Text Summarization

  2019cites this paper
• Segmentation of Rulemaking Documentsfor Public Notice-and-Comment Process Analysis

  2019cites this paper
• Weakly supervised sentiment analysis and opinion extraction

  2019influential citation
• Neural Text Summarization: A Critical Evaluation

  2019cites this paper
• Deep Communicating Agents for Abstractive Summarization

  2018cites this paper
• On the Abstractiveness of Neural Document Summarization

  2018cites this paper
• Automatic Pyramid Evaluation Exploiting EDU-based Extractive Reference Summaries

  2018cites this paper
• Chinese Discourse Segmentation Using Bilingual Discourse Commonality

  2018cites this paper
• Toward Fast and Accurate Neural Discourse Segmentation

  2018cites this paper
• Summarizing Opinions: Aspect Extraction Meets Sentiment Prediction and They Are Both Weakly Supervised

  2018cites this paper
• ODEL FOR A BSTRACTIVE S UMMARIZATION

  2018cites this paper
• Patterns of Argumentation Strategies across Topics

  2017cites this paper
• Summarizing Dialogic Arguments from Social Media

  2017cites this paper
• Recent advances in document summarization

  2017cites this paper
• Dialogic Arguments from Social Media

  2017cites this paper
• Multiple Instance Learning Networks for Fine-Grained Sentiment Analysis

  2017cites this paper
• A Deep Reinforced Model for Abstractive Summarization

  2017cites this paper
• Multi-Structured Models for Transforming and Aligning Text

  2015cites this paper
• ComputerSpeechandLanguage QBSUM: a Large-Scale Query-Based Document Summarization Dataset from Real-world Applications

  year unknowncites this paper