AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

LIFT: Latent Implicit Functions for Task- and Data-Agnostic Encoding

Amirhossein Kazerouni, Soroush Mehraban, Michael Brudno, Babak Taati

ICCV 2025 · Image generation › Autoregressive and latent generation · approach: Latent diffusion with learned tokenizers · 2 citations (OpenAlex)

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
add intermediate latent to global+local (ablation)LIFT · CelebA-HQ reconstructionTest PSNR: 39.62 -> 40.91Improved
add local latent to global-only (ablation)LIFT · CelebA-HQ reconstructionTest PSNR: 24.82 -> 39.62Improved
add hierarchical latent generator (HLG) (ablation)LIFT · CelebA-HQ reconstructionTest PSNR: 39.28 -> 40.91Improved
add hierarchical latent generator (HLG) (ablation)LIFT · CelebA-HQ reconstructionrFID: 2.70 -> 2.22Improved
increase Tinner from 1 to 3 (ablation)LIFT · CelebA-HQ reconstructionTest PSNR: 33.25 -> 40.17Improved
increase Tinner from 3 to 4 (ablation)LIFT · CelebA-HQ reconstructionTest PSNR: 40.17 -> 40.28No clear effect
double local latent spatial size from 4x4 to 8x8 (ablation)LIFT · CelebA-HQ reconstructionTest PSNR: 30.27 -> 40.42Improved
enlarge intermediate latent from 2x2 to 4x4 (ablation)LIFT · CelebA-HQ reconstructionTest PSNR: 40.42 -> 40.91Improved
double channels of all latents (ablation)LIFT · CelebA-HQ reconstructionTest PSNR: 40.91 -> 49.86Improved
increase P-MLP depth from 1x256 to 16x16 (narrower, deeper) (ablation)LIFT P-MLP · CelebA-HQ reconstructionTest PSNR: 40.25 -> 35.16Hurt
increase augmentations from 1 to 10 (ablation)LIFT · CIFAR-10 classificationTop-1 accuracy: 91.04 -> 95.47Improved
use LIFT multiscale latent instead of global Functa latentLIFT · CIFAR-10 classificationTop-1 accuracy: 68.30 -> 95.47Improved

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