Persuasiveness of Generated Free-Text Rationales in Subjective Decisions: A Case Study on Pairwise Argument Ranking

TL;DR

This study evaluates how effectively large language models generate persuasive free-text rationales for subjective tasks, specifically pairwise argument ranking, revealing that open-source models like Llama2-70B-chat outperform some GPT models in persuasiveness.

Contribution

It provides the first comprehensive analysis of rationale persuasiveness in subjective decision tasks, highlighting the potential of open-source LLMs and methods to enhance rationale quality.

Findings

Open-source Llama2-70B-chat outperforms GPT models in persuasiveness.

Rationale persuasiveness improves with prompting and self-refinement.

Generated rationales are highly persuasive in subjective argument ranking.

Abstract

Generating free-text rationales is among the emergent capabilities of Large Language Models (LLMs). These rationales have been found to enhance LLM performance across various NLP tasks. Recently, there has been growing interest in using these rationales to provide insights for various important downstream tasks. In this paper, we analyze generated free-text rationales in tasks with subjective answers, emphasizing the importance of rationalization in such scenarios. We focus on pairwise argument ranking, a highly subjective task with significant potential for real-world applications, such as debate assistance. We evaluate the persuasiveness of rationales generated by nine LLMs to support their subjective choices. Our findings suggest that open-source LLMs, particularly Llama2-70B-chat, are capable of providing highly persuasive rationalizations, surpassing even GPT models. Additionally, our experiments show that rationale persuasiveness can be improved by controlling its parameters through prompting or through self-refinement.