On the Effectiveness of LLM-Specific Fine-Tuning for Detecting AI-Generated Text

TL;DR

This paper evaluates the effectiveness of fine-tuning large language models specifically for detecting AI-generated text, demonstrating significant accuracy improvements over existing methods.

Contribution

It introduces large-scale corpora and novel fine-tuning strategies tailored for LLM detection, achieving state-of-the-art accuracy.

Findings

Best model achieves 99.6% token-level accuracy

Fine-tuning per LLM improves detection performance

New training paradigms outperform existing baselines

Abstract

The rapid progress of large language models has enabled the generation of text that closely resembles human writing, creating challenges for authenticity verification in education, publishing, and digital security. Detecting AI-generated text has therefore become a crucial technical and ethical issue. This paper presents a comprehensive study of AI-generated text detection based on large-scale corpora and novel training strategies. We introduce a 1-billion-token corpus of human-authored texts spanning multiple genres and a 1.9-billion-token corpus of AI-generated texts produced by prompting a variety of LLMs across diverse domains. Using these resources, we develop and evaluate numerous detection models and propose two novel training paradigms: Per LLM and Per LLM family fine-tuning. Across a 100-million-token benchmark covering 21 large language models, our best fine-tuned detector achieves up to $99.6\%$ token-level accuracy, substantially outperforming existing open-source baselines.