A Logical Fallacy-Informed Framework for Argument Generation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Mouchel, Luca, Paul, Debjit, Cui, Shaobo, West, Robert, Bosselut, Antoine, Faltings, Boi
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915271924514816
author Mouchel, Luca
Paul, Debjit
Cui, Shaobo
West, Robert
Bosselut, Antoine
Faltings, Boi
author_facet Mouchel, Luca
Paul, Debjit
Cui, Shaobo
West, Robert
Bosselut, Antoine
Faltings, Boi
contents Despite the remarkable performance of Large Language Models (LLMs) in natural language processing tasks, they still struggle with generating logically sound arguments, resulting in potential risks such as spreading misinformation. To address this issue, we introduce FIPO, a fallacy-informed framework that leverages preference optimization methods to steer LLMs toward logically sound arguments. FIPO includes a classification loss, to capture the fine-grained information on fallacy types. Our results on argumentation datasets show that our method reduces the fallacy errors by up to 17.5%. Furthermore, our human evaluation results indicate that the quality of the generated arguments by our method significantly outperforms the fine-tuned baselines, as well as other preference optimization methods, such as DPO. These findings highlight the importance of ensuring models are aware of logical fallacies for effective argument generation. Our code is available at github.com/lucamouchel/Logical-Fallacies.
format Preprint
id arxiv_https___arxiv_org_abs_2408_03618
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle A Logical Fallacy-Informed Framework for Argument Generation
Mouchel, Luca
Paul, Debjit
Cui, Shaobo
West, Robert
Bosselut, Antoine
Faltings, Boi
Computation and Language
Artificial Intelligence
Machine Learning
Despite the remarkable performance of Large Language Models (LLMs) in natural language processing tasks, they still struggle with generating logically sound arguments, resulting in potential risks such as spreading misinformation. To address this issue, we introduce FIPO, a fallacy-informed framework that leverages preference optimization methods to steer LLMs toward logically sound arguments. FIPO includes a classification loss, to capture the fine-grained information on fallacy types. Our results on argumentation datasets show that our method reduces the fallacy errors by up to 17.5%. Furthermore, our human evaluation results indicate that the quality of the generated arguments by our method significantly outperforms the fine-tuned baselines, as well as other preference optimization methods, such as DPO. These findings highlight the importance of ensuring models are aware of logical fallacies for effective argument generation. Our code is available at github.com/lucamouchel/Logical-Fallacies.
title A Logical Fallacy-Informed Framework for Argument Generation
topic Computation and Language
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2408.03618