Trinity-RFT: A General-Purpose and Unified Framework for Reinforcement Fine-Tuning of Large Language Models Paper • 2505.17826 • Published May 23, 2025 • 9
