Joint Modeling of Entities and Discourse Relations for Coherence Assessment

Wei Liu, Michael Strube
9/4/2025
cs.CL

No official code found. Best implementation: liuwei1206/EntyRelCoh (2 stars, last commit 2026-01-18).

Abstract

In linguistics, coherence can be achieved by different means, such as by maintaining reference to the same set of entities across sentences and by establishing discourse relations between them. However, most existing work on coherence modeling focuses exclusively on either entity features or discourse relation features, with little attention given to combining the two. In this study, we explore two methods for jointly modeling entities and discourse relations for coherence assessment. Experiments on three benchmark datasets show that integrating both types of features significantly enhances the performance of coherence models, highlighting the benefits of modeling both simultaneously for coherence evaluation.

View on arXivView PDF

Code Implementations(1)

liuwei1206/EntyRelCohCommunity74%

Code for the EMNLP2025 paper "Joint Modeling of Entities and Discourse Relations for Coherence Assessment".

20Shell, PythonAug 22, 20259 months ago

Similar Papers

A Novel Computational and Modeling Foundation for Automatic Coherence Assessment

83% match2023

CORDIAL: Can Multimodal Large Language Models Effectively Understand Coherence Relationships?

80% match1 repo2025

BBScore: A Brownian Bridge Based Metric for Assessing Text Coherence

80% match2023

Unlocking Structure Measuring: Introducing PDD, an Automatic Metric for Positional Discourse Coherence

78% match1 repo72024

Measuring Semantic Coherence of a Conversation

78% match1 repo182018

Synthetic Data Augmentation for Cross-domain Implicit Discourse Relation Recognition

78% match2025

Cite this paper

@article{liu2025joint,
  title  = {Joint Modeling of Entities and Discourse Relations for Coherence Assessment},
  author = {Wei Liu and Michael Strube},
  year   = {2025},
  eprint = {2509.04182},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2509.04182v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2509.04182.svg)](https://codeofpaper.com/paper/2509.04182)

Discussion