AI RESEARCH
Seeing What Matters: Visual Preference Policy Optimization for Visual Generation
arXiv CS.CV
•
ArXi:2511.18719v4 Announce Type: replace Reinforcement learning (RL) has become a powerful tool for post-