AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Debiasing Pretrained Generative Models by Uniformly Sampling Semantic Attributes

Walter Gerych, Kevin Hickey, Luke Buquicchio, Kavin Chandrasekaran, Abdulaziz Alajaji, Elke Rundensteiner, Emmanuel Agu

NeurIPS 2023 · Image generation › Fast diffusion sampling · approach: Distributional control of diffusion

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Apply Distribution Mapping to the generatorProgressive GAN · CelebA-HQ RaceKL divergence from uniform: 0.68 -> 0.41Improved
Apply Distribution Mapping to the latent diffusion modelLDM · CelebA-HQ AgeKL divergence from uniform: 0.411 -> 0.130Improved
Use Latent Editing instead of vanilla samplingLDM · CelebA-HQ AgeKL divergence from uniform: 0.411 -> 0.525Hurt
Use Latent Editing instead of vanilla samplingProgressive GAN · CelebA-HQ RaceKL divergence from uniform: 0.68 -> 0.62Improved
Use Polarity antimode sampling (rho=2) instead of standard samplingProgressive GAN · CelebA-HQ RaceKL divergence from uniform: 0.68 -> 0.49Improved
Use Polarity mode sampling (rho=-2) instead of standard samplingProgressive GAN · CelebA-HQ RaceKL divergence from uniform: 0.68 -> 0.57Improved
Use MaGNET instead of standard samplingProgressive GAN · CelebA-HQ RaceKL divergence from uniform: 0.68 -> 0.56Improved
Use Shifted Semantic Distribution instead of standard samplingProgressive GAN · CelebA-HQ RaceKL divergence from uniform: 0.68 -> 0.60Improved
Add bias correction (Algorithm 2) when building the Fair Noise dataset with a biased linear classifier (ablation)Fair Noise dataset construction · Shapes (circle/square)group density (ideal 0.5)Improved
Add bias correction (Algorithm 2) with a weakened biased classifier (ablation)Fair Noise dataset construction · UTKFace DCGAN, 10 age binsgroup density (ideal 0.1)Improved
Use Latent Editing with biased classifier rather than perfect classifier (ablation)Latent Editing · Shapesgroup density (ideal 0.5)Hurt
Use Distribution Mapping vs baselines on age debiasingDistribution Mapping · UTKFace DCGAN Ageage distribution uniformityMixed

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy