AI Paper Atlas

Research map / Benchmarks / LSUN-Church 256

LSUN-Church 256 leaderboard

10 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on LSUN-Church 256. Best FID: 1.59 (Projected GAN, NeurIPS 2021).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

LSUN-Church 256 · FID

10 papers, each with its best reported number for its own method; lower is better.

#FID ↓MethodPaperTopic
11.59Projected GANProjected GANs Converge Faster
NeurIPS 2021 · Quantitative Results table
GANs
21.72Ours (w/ ADA)
StyleGAN2 (F)
Ensembling Off-the-Shelf Models for GAN Training
CVPR 2022 · Table: Results on full-dataset
GANs
32.32StyleGAN2 w/ MTM
StyleGAN2
Learning Modulated Transformation in GANs
NeurIPS 2023 · Table 1
GANs
45.25Denoising Diffusion GANTackling the Generative Learning Trilemma with Denoising Diffusion GANs
ICLR 2022 · Table: Generative results on L
Fast diffusion sampling
55.5CREPSEfficient Scale-Invariant Generator With Column-Row Entangled Pixel Synthesis
CVPR 2023 · Table: FID comparison
Autoregressive and latent generation
67.03INRFlowINRFlow: Flow Matching for INRs in Ambient Space
ICML 2025 · Architecture ablation table
Point cloud classification
77.26Noise Corrector 200 + model 200 stepsFine-Tuning Diffusion Models via Intermediate Distribution Shaping
ICLR 2026 · Image generation table
Fast diffusion sampling
87.89Ours (L_simple)Denoising Diffusion Probabilistic Models
NeurIPS 2020 · Appendix table
Fast diffusion sampling
910.28GIRAFFE HD
GIRAFFE
GIRAFFE HD: A High-Resolution 3D-Aware Generative Model
CVPR 2022 · Table: 256² Resolution Image Q
Editing and style transfer
1011.252ndM (Ours)
DDPM pruned
2ndMatch: Finetuning Pruned Diffusion Models via Second-Order Jacobian Matching
CVPR 2026 · Table 1
Fast diffusion sampling

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.