most citedGenArena: How Can We Achieve Human-Aligned Evaluation for Visual Generation Tasks?

1 citations · 1 across the 3 of their papers we have counts for

collaborators
Showing cs.LGShow all

1 paper · 1 filter

cs.LG2026

Optimizing Visual Generative Models via Distribution-wise Rewards

Ruihang Li, Mengde Xu, Shuyang Gu +4

Conventional reinforcement learning strategies for visual generation typically employ sample-wise reward functions, yet this practice frequently results in reward hacking that degr…