Multitask Instruction-based Prompting for Fallacy Recognition

TL;DR

This paper introduces a multitask instruction-based prompting method using T5 to improve fallacy recognition across diverse datasets, genres, and fallacy types, outperforming models trained on specific datasets.

Contribution

It proposes a novel multitask prompting approach that handles dataset variability and recognizes 28 fallacy types, advancing fallacy detection methods.

Findings

Multitask prompting improves fallacy recognition accuracy.

The approach generalizes across domains and genres.

Model size and prompt choice significantly affect performance.

Abstract

Fallacies are used as seemingly valid arguments to support a position and persuade the audience about its validity. Recognizing fallacies is an intrinsically difficult task both for humans and machines. Moreover, a big challenge for computational models lies in the fact that fallacies are formulated differently across the datasets with differences in the input format (e.g., question-answer pair, sentence with fallacy fragment), genre (e.g., social media, dialogue, news), as well as types and number of fallacies (from 5 to 18 types per dataset). To move towards solving the fallacy recognition task, we approach these differences across datasets as multiple tasks and show how instruction-based prompting in a multitask setup based on the T5 model improves the results against approaches built for a specific dataset such as T5, BERT or GPT-3. We show the ability of this multitask prompting approach to recognize 28 unique fallacies across domains and genres and study the effect of model size and prompt choice by analyzing the per-class (i.e., fallacy type) results. Finally, we analyze the effect of annotation quality on model performance, and the feasibility of complementing this approach with external knowledge.