Research map / Image generation / Fast diffusion sampling
Improving Classifier-Free Guidance in Masked Diffusion: Low-Dim Theoretical Insights with High-Dim Impact
Kevin Rojas, Ye He, Chieh-Hsin Lai, Yuhta Takida, Yuki Mitsufuji, Molei Tao
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 |
|---|---|---|---|
| column-normalize the guided rate matrix (normalized guidance) (ablation) | MaskGIT with constant guidance, tau-leaping · ImageNet-256 (50K samples, 50 steps) | FID | Improved |
| use normalized guidance instead of unlocking or simple guidance at moderate w | MaskGIT · ImageNet-256 precision/recall | precision | Improved |
| add normalization to guidance (ablation) | Show-o and Meissonic · GenEval | prompt alignment score | Improved |
| add normalization to guidance (ablation) | LLaDA-8B-Instruct, Euler sampler without remasking · MATH-500 | accuracy | Improved |
| use increasing guidance schedules (Right Interval, Ramp-Up) (ablation) | MaskGIT · ImageNet-256 (10K samples) | FID | Improved |
| use Ramp-Up instead of Right Interval schedule (ablation) | MaskGIT · ImageNet-256 (10K samples) | FID | Improved |
| use decreasing guidance schedules (Left Interval, Ramp-Down) (ablation) | MaskGIT · ImageNet-256 (10K samples) | FID | Hurt |
| add normalized guidance in uniform diffusion | conditional QM9 uniform diffusion model · QM9 (ring-count conditioning) | validity, uniqueness, novelty | Improved |
Closest papers by meaning
- No Training, No Problem: Rethinking Classifier-Free Guidance for Diffusion Models (ICLR 2025 · similarity 0.94)
- Rectified Diffusion Guidance for Conditional Generation (CVPR 2025 · similarity 0.94)
- Eliminating Oversaturation and Artifacts of High Guidance Scales in Diffusion Models (ICLR 2025 · similarity 0.93)
- Elucidating the design space of classifier-guided diffusion generation (ICLR 2024 · similarity 0.92)
- CFG++: Manifold-constrained Classifier Free Guidance for Diffusion Models (ICLR 2025 · similarity 0.92)
- TFG: Unified Training-Free Guidance for Diffusion Models (NeurIPS 2024 · similarity 0.92)
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.
