Contrasting Linguistic Patterns in Human and LLM-Generated News Text

TL;DR

This study quantitatively compares human and LLM-generated news texts across linguistic, emotional, and sociolinguistic dimensions, revealing significant measurable differences and biases in language use and emotional expression.

Contribution

It provides a comprehensive, multi-dimensional analysis of linguistic and emotional differences between human and various LLM-generated texts, highlighting biases and stylistic variations.

Findings

Humans show more varied sentence lengths and vocabulary.

Humans exhibit stronger negative emotions and less joy.

LLMs tend to use more numbers, symbols, and pronouns.

Abstract

We conduct a quantitative analysis contrasting human-written English news text with comparable large language model (LLM) output from six different LLMs that cover three different families and four sizes in total. Our analysis spans several measurable linguistic dimensions, including morphological, syntactic, psychometric, and sociolinguistic aspects. The results reveal various measurable differences between human and AI-generated texts. Human texts exhibit more scattered sentence length distributions, more variety of vocabulary, a distinct use of dependency and constituent types, shorter constituents, and more optimized dependency distances. Humans tend to exhibit stronger negative emotions (such as fear and disgust) and less joy compared to text generated by LLMs, with the toxicity of these models increasing as their size grows. LLM outputs use more numbers, symbols and auxiliaries (suggesting objective language) than human texts, as well as more pronouns. The sexist bias prevalent in human text is also expressed by LLMs, and even magnified in all of them but one. Differences between LLMs and humans are larger than between LLMs.