MeanFlowNFT: Bringing Forward-Process RL to Average-Velocity Generators
- Published
- Source
- arXiv
- Paper number
- 646
- Field
- Computer Vision
- arXiv ID
- 2607.15273
Key points
- It is the first combination of forward-process RL with MeanFlow, which generates 1 to 4 steps by predicting interval-averaged velocity.
- It applies the DiffusionNFT objective through an instantaneous velocity predictor derived from MeanFlow identity, while sampling still uses average velocity.
- It theoretically inherits the policy improvement guarantee of DiffusionNFT.
- On SD3.5-M, it beats SOTA on 6 of 8 metrics, and on Wan2.1 video it uses 4 steps to beat LongCat-Video RL's 50-step score of 82.57 with 84.33 VBench.
- It shows consistent gains under test-time scaling as the number of steps increases.
Paper links
External research summaries. These are not HDATF publications or measured product results.