Research map / Probabilistic and scientific modeling / Variational and generative sampling
Flow Along the $K$-Amplitude for Generative Modeling
weitao Du, Jiasheng Tang, Shuning Chang, Yu Rong, Fan Wang, Shengchao Liu
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 |
|---|---|---|---|
| flow along K-amplitude (wavelet) instead of latent flow matching | LFM DiT L/2 · CelebA-HQ 256 unconditional | FID: 5.28 -> 4.99 | Improved |
| flow along K-amplitude (Fourier) instead of latent flow matching | LFM DiT L/2 · CelebA-HQ 256 unconditional | Recall: 0.48 -> 0.47 | No clear effect |
| use three scaling components instead of two (ablation) | K-Flow wavelet · LSUN Church 256 unconditional | FID: 5.37 -> 5.19 | Improved |
| K-Flow vs latent diffusion model | LDM · LSUN Church 256 unconditional | FID: 4.02 -> 5.19 | Hurt |
| use K-Flow Fourier with cfg=1.5 | LFM DiT L/2 · ImageNet class-conditional | FID: 2.78 -> 2.77 | No clear effect |
| K-Flow wavelet without guidance | LFM DiT L/2 · ImageNet class-conditional | FID: 14.0 -> 17.8 | Hurt |
| drop class label in high-scaling segments of generation (ablation) | K-Flow Fourier vs LFM · ImageNet class-conditional | class-dropping ratio: 3.25 -> 1.49 | Improved |
| flow along K-amplitude for molecular assembly | AssembleFlow · COD-Cluster17-All | PM (atom): 7.37 -> 7.23 | Improved |
Closest papers by meaning
- Align Your Flow: Scaling Continuous-Time Flow Map Distillation (NeurIPS 2025 · similarity 0.86)
- Expanding Flow Maps (NeurIPS 2026 · similarity 0.86)
- Self-Corrected Flow Distillation for Consistent One-Step and Few-Step Image Generation (AAAI 2025 · similarity 0.86)
- Contextual Flow Matching: Adaptive Step Selection in Flow Models for Efficient Visual Generation (NeurIPS 2026 · similarity 0.85)
- D-Flow: Differentiating through Flows for Controlled Generation (ICML 2024 · similarity 0.85)
- Scaling Rectified Flow Transformers for High-Resolution Image Synthesis (ICML 2024 · similarity 0.85)
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.
