Direct Preference Optimization
DPOAlso called: DPO
A preference optimization technique that directly trains a language model to prefer chosen responses over rejected ones without requiring an explicit reward model.
Explore more about Feedback Loops
Related terms
The process of optimizing an AI system to produce outputs that better match preferred responses, behaviors, or outcomes.
Preference DataData that records preferences between AI outputs or behaviors, typically used to train or optimize models toward preferred responses.
Reinforcement Learning from Human FeedbackRLHFA machine learning alignment technique that optimizes model behavior based on preferences gathered from human evaluators.
Reinforcement Learning from AI FeedbackRLAIFA fine-tuning method that optimizes model behavior using feedback generated by another AI system rather than human annotators.
Reward ModelingThe process of training a mathematical model to score AI outputs based on human or automated preferences.