Diffusion Reward Models
Reward models underpin the alignment of large language models, yet the dominant designs reduce each prompt--response pair to a point estimate or to a distribution from a fixed parametric family. This is at odds with human preference, which is inherently multimodal: the same response can be reasonably judged in many way...
Xiang-Yang Wang, Bing-Xiang He, Ze-Yuan Liu et al.
· 0 citations