Research map / Image generation / Super-resolution
Trust but Verify: Adaptive Conditioning for Reference-Based Diffusion Super-Resolution via Implicit Reference Correlation Modeling
Yuan Wang, Yuhao Wan, Siming Zheng, Bo Li, Qibin Hou, Peng-Tao Jiang
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 |
|---|---|---|---|
| add AICG gating to vanilla reference attention (ablation) | Ada-RefSR · WRSR | PSNR: 21.9508 -> 21.9722 | Improved |
| add AICG gating to vanilla reference attention (ablation) | Ada-RefSR · WRSR | CLIPIQA: 0.6471 -> 0.6612 | Improved |
| add AICG gating to vanilla reference attention (ablation) | Ada-RefSR · Face | PSNR: 27.0795 -> 27.1271 | Improved |
| add AICG gating to vanilla reference attention (ablation) | Ada-RefSR · Face | SSIM: 0.7495 -> 0.7523 | Improved |
| use AICG instead of a global learnable weight (ablation) | Ada-RefSR · WRSR | PSNR: 21.6267 -> 21.9722 | Improved |
| use AICG instead of ReFIR spatial gating (ablation) | Ada-RefSR · WRSR | PSNR: 21.7753 -> 21.9722 | Improved |
| add ReFIR spatial gating to vanilla reference attention (ablation) | Ada-RefSR · WRSR | PSNR: 21.9508 -> 21.7753 | Hurt |
| add global learnable weight to vanilla reference attention (ablation) | Ada-RefSR · WRSR | PSNR: 21.9508 -> 21.6267 | Hurt |
| add ReFIR spatial gating to vanilla reference attention (ablation) | Ada-RefSR · Face | CLIPIQA: 0.6444 -> 0.6483 | Mixed |
| add reference attention with AICG to S3Diff | Ada-RefSR · Face | PSNR: 26.6881 -> 27.1271 | Improved |
| add reference attention with AICG to S3Diff | Ada-RefSR · Bird | FID: 60.9452 -> 36.4249 | Improved |
| use implicit AICG gating instead of explicit ReFIR correlation | Ada-RefSR · computational overhead | added cost (%): 16.0% -> 0.13% | Improved |
Closest papers by meaning
- RAP-SR: RestorAtion Prior Enhancement in Diffusion Models for Realistic Image Super-Resolution (AAAI 2025 · similarity 0.89)
- TSD-SR: One-Step Diffusion with Target Score Distillation for Real-World Image Super-Resolution (CVPR 2025 · similarity 0.89)
- DREAM: Diffusion Rectification and Estimation-Adaptive Models (CVPR 2024 · similarity 0.88)
- Pixel-level and Semantic-level Adjustable Super-resolution: A Dual-LoRA Approach (CVPR 2025 · similarity 0.87)
- Building Bridges across Spatial and Temporal Resolutions: Reference-Based Super-Resolution via Change Priors and Conditional Diffusion Model (CVPR 2024 · similarity 0.87)
- SinSR: Diffusion-Based Image Super-Resolution in a Single Step (CVPR 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.
