Research map / Image generation / Fast diffusion sampling
Synthesize-It-Classifier: Learning a Generative Classifier Through Recurrent Self-Analysis
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 |
|---|---|---|---|
| train with VRM rather than ERM (ablation) | STIC · LSUN | FID: 35.92 -> 13.32 | Helped |
| train with VRM rather than ERM (ablation) | STIC · CIFAR10 | FID: 48.17 -> 12.81 | Helped |
| train with VRM rather than ERM (ablation) | STIC · ImageNet | FID: 38.27 -> 15.01 | Helped |
| use attentive feature-space synthesis (ablation) | STIC · LSUN | FID: 13.32 -> 9.01 | Helped |
| use attentive feature-space synthesis (ablation) | STIC · CIFAR10 | FID: 12.81 -> 11.71 | Helped |
| use attentive feature-space synthesis (ablation) | STIC · ImageNet | FID: 15.01 -> 10.38 | Helped |
| replace the standard classifier with a class-conditional score classifier (ablation) | STIC · LSUN | FID: 13.32 -> 8.82 | Helped |
| replace the standard classifier with a class-conditional score classifier (ablation) | STIC · CIFAR10 | FID: 12.81 -> 9.99 | Helped |
| replace the standard classifier with a class-conditional score classifier (ablation) | STIC · ImageNet | FID: 15.01 -> 8.83 | Helped |
| use shallow conv1'-conv20' Gram layers instead of deeper conv21'-conv28' layers (ablation) | STIC · ImageNet | FID: 15.01 -> 28 | Hurt |
| use all conv1'-conv28' Gram layers instead of deeper conv21'-conv28' layers (ablation) | STIC · ImageNet | FID: 15.01 -> 30 | Hurt |
| use STIC rather than BigGAN | image generation · LSUN | Cls_R: 88 -> 96 | Helped |
Closest papers by meaning
- Real-Fake: Effective Training Data Synthesis Through Distribution Matching (ICLR 2024 · similarity 0.87)
- Iterative Self-Training with Class-Aware Text-to-Image Synthesis for Visual Task Learning (AAAI 2025 · similarity 0.86)
- Leverage Class-Specific Accuracy to Guide Data Generation for Improving Image Classification (ICML 2024 · similarity 0.85)
- Provably Improving Generalization of Few-shot models with Synthetic Data (ICML 2025 · similarity 0.85)
- Enhance Image Classification via Inter-Class Image Mixup with Diffusion Model (CVPR 2024 · similarity 0.84)
- Generative Classifiers Avoid Shortcut Solutions (ICLR 2025 · similarity 0.84)
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.