GARDO: Reinforcing Diffusion Models without Reward Hacking Paper • 2512.24138 • Published 10 days ago • 28
GRPO-Guard: Mitigating Implicit Over-Optimization in Flow Matching via Regulated Clipping Paper • 2510.22319 • Published Oct 25, 2025 • 2
Scaling Image and Video Generation via Test-Time Evolutionary Search Paper • 2505.17618 • Published May 23, 2025 • 41