Research map / Image generation / GANs
Your GAN is Secretly an Energy-based Model and You Should Use Discriminator Driven Latent Sampling
Read the paperCheck 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 DDLS latent sampling | SNGAN · CIFAR-10 | Inception Score: 8.22 -> 9.09 | Helped |
| add DDLS latent sampling | SNGAN · CIFAR-10 | FID: 21.7 -> 15.42 | Helped |
| add DDLS latent sampling | SNGAN · ImageNet | Inception Score: 36.8 -> 40.2 | Helped |
| use DDLS instead of MH-GAN | DCGAN · CIFAR-10 | Inception Score: 3.379 -> 3.681 | Helped |
| use DDLS instead of MH-GAN | WGAN · CelebA | Inception Score: 2.889 -> 3.093 | Helped |
| use DDLS instead of DOT | ResNet-SAGAN / SNGAN · CIFAR-10 | Inception Score: 8.50 -> 9.09 | Helped |
| add DDLS to generator | WGAN · 25-Gaussians | EMD: 0.043 -> 0.036 | Helped |
| use DDLS instead of DOT | WGAN · Swiss Roll | EMD: 0.020 -> 0.020 | No clear effect |
| use DDLS instead of DRS | WGAN · 25-Gaussians | % high-quality samples: 90 -> 98 | Helped |
Closest papers by meaning
- Exploring and Exploiting Hubness Priors for High-Quality GAN Latent Sampling (ICML 2022 · similarity 0.89)
- Discriminator Rejection Sampling (ICLR 2019 · similarity 0.88)
- MCGAN: Enhancing GAN Training with Regression-Based Generator Loss (AAAI 2025 · similarity 0.88)
- Improving the Training of Data-Efficient GANs via Quality Aware Dynamic Discriminator Rejection Sampling (CVPR 2025 · similarity 0.88)
- Direct Discriminative Optimization: Your Likelihood-Based Visual Generative Model is Secretly a GAN Discriminator (ICML 2025 · similarity 0.87)
- The GAN is dead; long live the GAN! A Modern GAN Baseline (NeurIPS 2024 · similarity 0.87)
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.