SemEval-2023 Task 4: ValueEval

Identification of Human Values Behind Arguments

verfasst von

Johannes Kiesel, Milad Alshomary, Nailia Mirzakhmedova, Maximilian Heinrich, Nicolas Handke, Henning Wachsmuth, Benno Stein

Abstract

Argumentation is ubiquitous in natural language communication, from politics and media to everyday work and private life. Many arguments derive their persuasive power from human values, such as self-directed thought or tolerance, albeit often implicitly. These values are key to understanding the semantics of arguments, as they are generally accepted as justifications for why a particular option is ethically desirable. Can automated systems uncover the values on which an argument draws? To answer this question, 39 teams submitted runs to ValueEval’23. Using a multi-sourced dataset of over 9K arguments, the systems achieved F

₁-scores up to 0.87 (nature) and over 0.70 for three more of 20 universal value categories. However, many challenges remain, as evidenced by the low peak F

₁-score of 0.39 for stimulation, hedonism, face, and humility.

Organisationseinheit(en)

Fachgebiet Maschinelle Sprachverarbeitung

Externe Organisation(en)

Bauhaus-Universität Weimar
Universität Leipzig

Typ

Aufsatz in Konferenzband

Seiten

2287-2303

Anzahl der Seiten

17

Publikationsdatum

07.2023

Publikationsstatus

Veröffentlicht

Peer-reviewed

Ja

ASJC Scopus Sachgebiete

Theoretische Informatik, Angewandte Informatik, Theoretische Informatik und Mathematik

Elektronische Version(en)

https://doi.org/10.18653/V1/2023.SEMEVAL-1.313 (Zugang: Offen)