AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Multi-Label Conditional Generation From Pre-Trained Models

Magdalena Proszewska, Maciej Wołczyk, Maciej Zięba, Patryk Wielopolski, Łukasz Maziarka, Marek Śmieja

TPAMI 2024 · Image generation › Autoregressive and latent generation · approach: Generative latent space tooling

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
use PluGeN instead of StyleFlowStyleGAN · FFHQ attribute modificationaverage accuracy: 0.86 -> 0.90Improved
use PluGeN instead of StyleFlowStyleGAN · FFHQ attribute modificationFID: 32.59 -> 46.51Hurt
use PluGeN instead of cFlowVAE · CelebA attribute manipulationaverage FID: 39.68 -> 28.07Improved
use PluGeN instead of MSPVAE · CelebA attribute manipulationaverage accuracy: 0.70 -> 0.69No clear effect
use PluGeN instead of MSP for rare combination female x beardVAE · CelebA attribute manipulationaccuracy: 0.33 -> 0.59Improved
use PluGeN instead of cFlowVAE · CelebA generation, independently sampled attributesF1: 0.29 -> 0.44Improved
use PluGeN instead of cVAEVAE · CelebA generation, test-set attribute combinationsF1: 0.59 -> 0.69Improved
replace NICE with continuous normalizing flow (ablation)PluGeN · CelebA generation, test-set attribute combinationsFID: 72.72 -> 68.96Improved
replace NICE with continuous normalizing flow (ablation)PluGeN · CelebA generation, test-set attribute combinationsF1: 0.69 -> 0.63Hurt
replace NICE with continuous normalizing flow (ablation)PluGeN · CelebA generation, independently sampled attributesF1: 0.44 -> 0.41Hurt
increase beta-VAE beta from 1 to 16 (ablation)PluGeN · CelebA beta-VAE backboneFID: 55.11 -> 110.46Hurt
lower beta-VAE beta from 1 to 0.5 (ablation)PluGeN · CelebA beta-VAE backboneF1: 0.66 -> 0.45Hurt

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