Min-K%++: Improved Baseline for Detecting Pre-Training Data from Large Language Models Paper • 2404.02936 • Published Apr 3, 2024 • 3 • 1