A Lightweight Approach to Detection of AI-Generated Texts Using Stylometric Features

TL;DR

This paper presents NEULIF, a lightweight, computationally efficient method for detecting AI-generated texts using stylometric and readability features, achieving high accuracy comparable to complex models.

Contribution

Introduces NEULIF, a novel lightweight detection approach that combines stylometric features with simple classifiers, outperforming existing lightweight methods in accuracy and efficiency.

Findings

Achieves 97% accuracy with CNN on Kaggle dataset

Models are significantly smaller and faster than transformer-based methods

High potential for cross-language and real-time applications

Abstract

A growing number of AI-generated texts raise serious concerns. Most existing approaches to AI-generated text detection rely on fine-tuning large transformer models or building ensembles, which are computationally expensive and often provide limited generalization across domains. Existing lightweight alternatives achieved significantly lower accuracy on large datasets. We introduce NEULIF, a lightweight approach that achieves best performance in the lightweight detector class, that does not require extensive computational power and provides high detection accuracy. In our approach, a text is first decomposed into stylometric and readability features which are then used for classification by a compact Convolutional Neural Network (CNN) or Random Forest (RF). Evaluated and tested on the Kaggle AI vs. Human corpus, our models achieve 97% accuracy (~ 0.95 F1) for CNN and 95% accuracy (~ 0.94 F1) for the Random Forest, demonstrating high precision and recall, with ROC-AUC scores of 99.5% and 95%, respectively. The CNN (~ 25 MB) and Random Forest (~ 10.6 MB) models are orders of magnitude smaller than transformer-based ensembles and can be run efficiently on standard CPU devices, without sacrificing accuracy. This study also highlights the potential of such models for broader applications across languages, domains, and streaming contexts, showing that simplicity, when guided by structural insights, can rival complexity in AI-generated content detection.