Aligning protein-generative models to experimental fitness with ProteinDPO
Generative model alignment Generative AI applications are now routinely fine-tuned with feedback-based RL during post-training,
ProteinDPO