Research map / Probabilistic and scientific modeling / Variational and generative sampling
Deep Generative Modeling on Limited Data with Regularization by Nontransferable Pre-trained Models
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 Reg-DGM regularization | StyleGAN2 · FFHQ 5k | median FID: 52.71 -> 37.77 | Helped |
| Add Reg-DGM regularization | ADA · FFHQ 5k | median FID: 12.64 -> 11.95 | Helped |
| Add Reg-DGM regularization | ADA + APA · FFHQ 5k | median FID: 8.84 -> 8.02 | Helped |
| Use an ImageNet ResNet-18 feature extractor rather than no regularization | StyleGAN2 · FFHQ 5k | median FID: 52.71 -> 37.77 | Helped |
| Replace feature-distance energy with data-independent entropy regularization (ablation) | StyleGAN2 · FFHQ 5k | FID: 52.71 -> 50.87 | No clear effect |
| Replace expected feature-distance energy with averaged-feature matching (ablation) | StyleGAN2 · FFHQ 5k | FID: 52.71 -> 32.65 | Helped |
| Replace ImageNet ResNet-18 with CLIP image encoder (ablation) | Reg-StyleGAN2 · FFHQ 5k | FID: 52.71 -> 40.98 | Helped |
| Replace ImageNet ResNet-18 with FaceNet face recognizer (ablation) | Reg-StyleGAN2 · FFHQ 5k | FID: 52.71 -> 38.80 | Helped |
Closest papers by meaning
- Generalizing Dataset Distillation via Deep Generative Prior (CVPR 2023 · similarity 0.88)
- Statistically Optimal Generative Modeling with Maximum Deviation from the Empirical Distribution (ICML 2024 · similarity 0.88)
- How Realistic Is Your Synthetic Data? Constraining Deep Generative Models for Tabular Data (ICLR 2024 · similarity 0.87)
- Generative Marginalization Models (ICML 2024 · similarity 0.87)
- Unified Generation, Reconstruction, and Representation: Generalized Diffusion with Adaptive Latent Encoding-Decoding (ICML 2024 · similarity 0.87)
- A Likelihood Based Approach to Distribution Regression Using Conditional Deep Generative Models (ICML 2025 · 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.