The Anatomy of Speech Persuasion: Linguistic Shifts in LLM-Modified Speeches

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Barkar, Alisa, Chollet, Mathieu, Labeau, Matthieu, Biancardi, Beatrice, Clavel, Chloe
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866913907444023296
author Barkar, Alisa
Chollet, Mathieu
Labeau, Matthieu
Biancardi, Beatrice
Clavel, Chloe
author_facet Barkar, Alisa
Chollet, Mathieu
Labeau, Matthieu
Biancardi, Beatrice
Clavel, Chloe
contents This study examines how large language models understand the concept of persuasiveness in public speaking by modifying speech transcripts from PhD candidates in the "Ma These en 180 Secondes" competition, using the 3MT French dataset. Our contributions include a novel methodology and an interpretable textual feature set integrating rhetorical devices and discourse markers. We prompt GPT-4o to enhance or diminish persuasiveness and analyze linguistic shifts between original and generated speech in terms of the new features. Results indicate that GPT-4o applies systematic stylistic modifications rather than optimizing persuasiveness in a human-like manner. Notably, it manipulates emotional lexicon and syntactic structures (such as interrogative and exclamatory clauses) to amplify rhetorical impact.
format Preprint
id arxiv_https___arxiv_org_abs_2506_18621
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle The Anatomy of Speech Persuasion: Linguistic Shifts in LLM-Modified Speeches
Barkar, Alisa
Chollet, Mathieu
Labeau, Matthieu
Biancardi, Beatrice
Clavel, Chloe
Computation and Language
This study examines how large language models understand the concept of persuasiveness in public speaking by modifying speech transcripts from PhD candidates in the "Ma These en 180 Secondes" competition, using the 3MT French dataset. Our contributions include a novel methodology and an interpretable textual feature set integrating rhetorical devices and discourse markers. We prompt GPT-4o to enhance or diminish persuasiveness and analyze linguistic shifts between original and generated speech in terms of the new features. Results indicate that GPT-4o applies systematic stylistic modifications rather than optimizing persuasiveness in a human-like manner. Notably, it manipulates emotional lexicon and syntactic structures (such as interrogative and exclamatory clauses) to amplify rhetorical impact.
title The Anatomy of Speech Persuasion: Linguistic Shifts in LLM-Modified Speeches
topic Computation and Language
url https://arxiv.org/abs/2506.18621