#flow-matching
- Asymmetric Flow Models: SOTA 1.57 FID on ImageNet via Rank-Asymmetric Velocity Parameterization Stanford University research
- MrFlow: Training-Free 10-25x Speedup for Flow-Matching Text-to-Image Models Beihang University research
- Flow-DPPO: Principled RL Alignment for Flow Matching Image and Video Models Tencent Hunyuan research
- Self-OPD: On-Policy Distillation for Flow Matching Models without Teacher research
- DanceOPD: On-Policy Generative Field Distillation for Unified Image Generation ByteDance Seed research