Research map / Image generation / GANs
Manifold Learning Benefits GANs
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| Add LCSA to OmniGAN | OmniGAN · CIFAR-10, d'=512 | tFID: 6.88 -> 3.36 | Helped |
| Add LCSA to OmniGAN | OmniGAN · CIFAR-100, d'=512 | tFID: 9.13 -> 5.22 | Helped |
| Add LCSA to OmniGAN | OmniGAN · ImageNet 64×64 | IS: 70.59 -> 77.45 | Helped |
| Add LCSA to OmniGAN | OmniGAN · ImageNet 128×128 | tFID: 8.30 -> 4.83 | Helped |
| Add LCSA to StyleGAN2 | StyleGAN2 · FFHQ 256×256, 140k | FID: 3.71 -> 3.32 | Helped |
| Add proximity loss (ablation) | OmniGAN+LCSA · CIFAR-10, d'=512 | tFID: 5.15 -> 3.29 | Helped |
| Add proximity loss (ablation) | OmniGAN+LCSA · CIFAR-100, d'=512 | tFID: 8.12 -> 5.12 | Helped |
| Add adaptive mixing input (ablation) | OmniGAN+LCSA · CIFAR-10, d'=512 | tFID: 4.77 -> 3.29 | Helped |
| Add adaptive mixing input (ablation) | OmniGAN+LCSA · CIFAR-100, d'=512 | tFID: 5.54 -> 5.12 | Helped |
| Use learned dictionary updates instead of EMA (ablation) | OmniGAN+LCSA · CIFAR-100, d'=512 | tFID: 5.63 -> 5.12 | Helped |
| Use adaptive rather than fixed β and γ (ablation) | OmniGAN+LCSA · CIFAR-10, d'=512 | tFID: 4.06 -> 3.29 | Helped |
| Use adaptive rather than fixed β and γ (ablation) | OmniGAN+LCSA · CIFAR-100, d'=512 | tFID: 5.46 -> 5.12 | Helped |
Closest papers by meaning
- Global Versus Localized Generative Adversarial Nets (CVPR 2018 · similarity 0.91)
- Improving GAN with Neighbors Embedding and Gradient Matching (AAAI 2019 · similarity 0.89)
- Disconnected Manifold Learning for Generative Adversarial Networks (NeurIPS 2018 · similarity 0.89)
- A Geometric Analysis of Deep Generative Image Models and Its Applications (ICLR 2021 · similarity 0.89)
- What Secrets Do Your Manifolds Hold? Understanding the Local Geometry of Generative Models (ICLR 2025 · similarity 0.86)
- SMaRt: Improving GANs with Score Matching Regularity (ICML 2024 · similarity 0.86)
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.