What does “DPO (Direct Preference Optimization)” mean?

A training method that uses human preference signals to shape model behavior.

Training steps applied after pretraining, such as instruction tuning, preference optimization, and safety tuning.

Training approaches that let a model keep learning from new data without forgetting prior knowledge.

Continuing training on domain-specific data to adapt a pre-trained model to a specific task.

An optimization method that updates parameters in the direction that reduces error.

Fine-tuning a model on instruction-response pairs to improve task following.