AI RESEARCH
Alignment Imprint: Zero-Shot AI-Generated Text Detection via Provable Preference Discrepancy
arXiv CS.AI
•
ArXi:2604.16923v1 Announce Type: new Detecting AI-generated text is an important but challenging problem. Existing likelihood-based detection methods are often sensitive to content complexity and may exhibit unstable performance. In this paper, our key insight is that modern Large Language Models (LLMs) undergo alignment (including fine-tuning and preference tuning), leaving a measurable distributional imprint.