ORPO

Instruction tuning and preference alignment are essential techniques for adapting Large Language Models (LLMs) to specific tasks. Traditionally, this involves a multi-stage process: 1/ Supervised Fine-Tuning (SFT) on instructions to adapt the model to the target domain, followed by 2/ preference alignment methods like Reinforcement Learning with Human Feedback (RLHF) or Direct Preference Optimization (DPO) to increase the likelihood of generating preferred responses over rejected ones.