Belief-based Generation of Argumentative Claims

verfasst von

Milad Alshomary, Wei Fan Chen, Timon Gurcke, Henning Wachsmuth

Abstract

When engaging in argumentative discourse, skilled human debaters tailor claims to the audience's beliefs to construct effective arguments. Recently, the field of computational argumentation witnessed extensive effort to address the automatic generation of arguments. However, existing approaches do not perform any audience-specific adaptation. In this work, we aim to bridge this gap by studying the task of belief-based claim generation: Given a controversial topic and a set of beliefs, generate an argumentative claim tailored to the beliefs. To tackle this task, we model the people's prior beliefs through their stances on controversial topics and extend state-of-the-art text generation models to generate claims conditioned on the beliefs. Our automatic evaluation confirms the ability of our approach to adapt claims to a set of given beliefs. In a manual study, we also evaluate the generated claims in terms of informativeness and their likelihood to be uttered by someone with a respective belief. Our results reveal the limitations of modeling users' beliefs based on their stances. Still, they demonstrate the potential of encoding beliefs into argumentative texts, laying the ground for future exploration of audience reach.

Externe Organisation(en)

Universität Paderborn

Typ

Aufsatz in Konferenzband

Seiten

224-233

Anzahl der Seiten

10

Publikationsdatum

04.2021

Publikationsstatus

Veröffentlicht

Peer-reviewed

Ja

ASJC Scopus Sachgebiete

Software, Theoretische Informatik und Mathematik, Linguistik und Sprache

Elektronische Version(en)

https://doi.org/10.48550/arXiv.2101.09765 (Zugang: Offen)
https://doi.org/10.18653/v1/2021.eacl-main.17 (Zugang: Offen)