Fast-DetectGPT

anchor

Findings

• IC-335GPT-4's detection performance as a scoring model is highly sensitive to the prompt, varying from 0.7289 to 0.9682 AUROC, far more than GPT-3.5 or Babbage [primary]
• IC-336Larger proprietary LLMs (GPT-3.5 175B) are more effective universal text detectors than smaller models (Babbage 1.3B, GPT-Neo-2.7B), contradicting prior findings that smaller models are better [primary]
• IC-337GPT-3.5-based detection accuracy drops substantially for Russian text (0.8555 AUROC) compared to near-perfect scores for Urdu, Indonesian, and Arabic, suggesting under-training on Russian [primary]