Optimal Budgeted Rejection Sampling for Generative Models

Open paper
Benchmark Model Rank Results
image-generation-on-celeba-64x64BigGAN-OBRS–FID: 3.74Precision: 0.74Recall: 0.65
image-generation-on-imagenet-128x128BigGAN-OBRS–FID: 11.65Precision: 0.27Recall: 0.46

Rank counts only results with a code link.