SMLT-MUGC: Small, Medium, and Large Texts -- Machine versus User-Generated Content Detection and Comparison

TL;DR

This study evaluates the effectiveness of machine learning models in detecting texts generated by large language models across different text lengths and analyzes their linguistic and moral characteristics.

Contribution

It provides a comprehensive comparison of detection accuracy for LLM-generated texts of varying sizes and explores the linguistic and moral differences between human and machine texts.

Findings

Detection accuracy is high for smaller LLMs (96%+).

Detection becomes more challenging for very large LLMs (74%).

Machine texts have higher readability and similar moral judgments but differ in personality traits.

Abstract

Large language models (LLMs) have gained significant attention due to their ability to mimic human language. Identifying texts generated by LLMs is crucial for understanding their capabilities and mitigating potential consequences. This paper analyzes datasets of varying text lengths: small, medium, and large. We compare the performance of machine learning algorithms on four datasets: (1) small (tweets from Election, FIFA, and Game of Thrones), (2) medium (Wikipedia introductions and PubMed abstracts), and (3) large (OpenAI web text dataset). Our results indicate that LLMs with very large parameters (such as the XL-1542 variant of GPT2 with 1542 million parameters) were harder (74%) to detect using traditional machine learning methods. However, detecting texts of varying lengths from LLMs with smaller parameters (762 million or less) can be done with high accuracy (96% and above). We examine the characteristics of human and machine-generated texts across multiple dimensions, including linguistics, personality, sentiment, bias, and morality. Our findings indicate that machine-generated texts generally have higher readability and closely mimic human moral judgments but differ in personality traits. SVM and Voting Classifier (VC) models consistently achieve high performance across most datasets, while Decision Tree (DT) models show the lowest performance. Model performance drops when dealing with rephrased texts, particularly shorter texts like tweets. This study underscores the challenges and importance of detecting LLM-generated texts and suggests directions for future research to improve detection methods and understand the nuanced capabilities of LLMs.