Will It Blend? Mixing Training Paradigms & Prompting for Argument Quality Prediction

TL;DR

This paper explores combining different training methods and prompt engineering with GPT-3 to improve argument quality prediction, demonstrating that mixed approaches outperform single models in various aspects.

Contribution

It introduces a novel combination of training paradigms and prompt engineering techniques for argument quality prediction using large language models.

Findings

Mixed training paradigms outperform single models.

Prompt engineering with GPT-3 excels in argument validity prediction.

Multi-paradigm training best estimates argument novelty.

Abstract

This paper describes our contributions to the Shared Task of the 9th Workshop on Argument Mining (2022). Our approach uses Large Language Models for the task of Argument Quality Prediction. We perform prompt engineering using GPT-3, and also investigate the training paradigms multi-task learning, contrastive learning, and intermediate-task training. We find that a mixed prediction setup outperforms single models. Prompting GPT-3 works best for predicting argument validity, and argument novelty is best estimated by a model trained using all three training paradigms.