Research map / Image generation / Fast diffusion sampling
Omegance: A Single Parameter for Various Granularities in Diffusion-Based Synthesis
Xinyu Hou, Zongsheng Yue, Xiaoming Li, Chen Change Loy
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| apply Omegance varpi=6.0 (detail suppression) to SDXL (ablation) | SDXL · DiffusionDB 1000 prompts | FID: 162.18 -> 157.47 | Improved |
| apply Omegance varpi=6.0 to SDXL (ablation) | SDXL · DiffusionDB 1000 prompts | IS: 13.23 -> 13.82 | Improved |
| apply Omegance varpi=-6.0 (detail enhancement) to SDXL (ablation) | SDXL · DiffusionDB 1000 prompts | PickScore: 0.1468 -> 0.1601 | Improved |
| apply Omegance varpi=-6.0 to SDXL (ablation) | SDXL · DiffusionDB 1000 prompts | FID: 162.18 -> 170.52 | Hurt |
| apply Omegance varpi=6.0 to SDXL (ablation) | SDXL · DiffusionDB 1000 prompts | PickScore: 0.1468 -> 0.1149 | Hurt |
| add FreeU to SDXL | SDXL · DiffusionDB 1000 prompts | FID: 162.18 -> 167.22 | Hurt |
| use cosine scheduler instead of default | SDXL · DiffusionDB 1000 prompts | Q-Align: 4.68 -> 2.88 | Hurt |
| use rescaled scheduler instead of default | SDXL · DiffusionDB 1000 prompts | CLIP score: 32.88 -> 28.80 | Hurt |
| apply Omegance varpi=-6.0 to SDXL (ablation) | SDXL · DiffusionDB 1000 prompts | CLIP score: 32.88 -> 32.81 | No clear effect |
| increase omega globally (varpi=6.0) (ablation) | SDXL · DiffusionDB 1000 prompts | high-frequency energy: 1159.4 -> 955.2 | Improved |
| decrease omega globally (varpi=-6.0) (ablation) | SDXL · DiffusionDB 1000 prompts | high-frequency energy: 1159.4 -> 1680.1 | Improved |
| omega schedule EXP1 (early and late stage enhancement) (ablation) | SDXL · DiffusionDB 1000 prompts | SSIM: 1.0 -> 0.7087 | Mixed |
Closest papers by meaning
- Relay Diffusion: Unifying diffusion process across resolutions for image synthesis (ICLR 2024 · similarity 0.88)
- Faster Diffusion: Rethinking the Role of the Encoder for Diffusion Model Inference (NeurIPS 2024 · similarity 0.87)
- A Simple Early Exiting Framework for Accelerated Sampling in Diffusion Models (ICML 2024 · similarity 0.87)
- Simple and Fast Distillation of Diffusion Models (NeurIPS 2024 · similarity 0.87)
- Diffusion Model is Effectively Its Own Teacher (CVPR 2025 · similarity 0.87)
- EVODiff: Entropy-aware Variance Optimized Diffusion Inference (NeurIPS 2025 · 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.
