Measuring and Benchmarking Large Language Models' Capabilities to Generate Persuasive Language

TL;DR

This paper evaluates the ability of large language models to generate persuasive language across various domains, introducing a new dataset and benchmarking methods to measure and analyze their persuasive capabilities.

Contribution

It presents Persuasive-Pairs, a novel dataset for benchmarking persuasive language in LLMs, and analyzes how different prompts influence persuasive output across models.

Findings

LLMs can significantly alter persuasive language based on prompts.

The Persuasive-Pairs dataset enables effective benchmarking of LLMs' persuasive abilities.

Prompt engineering impacts the persuasiveness of generated text.

Abstract

We are exposed to much information trying to influence us, such as teaser messages, debates, politically framed news, and propaganda - all of which use persuasive language. With the recent interest in Large Language Models (LLMs), we study the ability of LLMs to produce persuasive text. As opposed to prior work which focuses on particular domains or types of persuasion, we conduct a general study across various domains to measure and benchmark to what degree LLMs produce persuasive language - both when explicitly instructed to rewrite text to be more or less persuasive and when only instructed to paraphrase. We construct the new dataset Persuasive-Pairs of pairs of a short text and its rewrite by an LLM to amplify or diminish persuasive language. We multi-annotate the pairs on a relative scale for persuasive language: a valuable resource in itself, and for training a regression model to score and benchmark persuasive language, including for new LLMs across domains. In our analysis, we find that different 'personas' in LLaMA3's system prompt change persuasive language substantially, even when only instructed to paraphrase.