Preprint
Aug 2026
Latent Reward Registers for Diffusion Preference Alignment
This work proposes Latent Reward Registers, a mechanism that estimates terminal preference directly from intermediate noisy latents, and achieves significant reward improvement with a favorable reward-quality balance against training-free baselines.
Yuan-Shen Guan, Zipeng Feng, Cheng-Ru Song et al.
· 0 citations