Stitched Value Model for Diffusion Alignment

Hyojun Go¹  ·  Hyungjin Chung  ·  Prune Truong²  ·  Goutam Bhat²  ·  Li Mi¹  ·  Zhaochong An³
Zixiang Zhao¹  ·  Dominik Narnhofer¹
Serge Belongie³  ·  Federico Tombari²  ·  Konrad Schindler¹

¹ETH Zürich  ·  ²Google  ·  ³University of Copenhagen

Project page  |  Paper (PDF)  |  arXiv

---

StitchVM turns any pretrained pixel reward model (CLIP, HPSv2, Aesthetic Predictor, …) into a value model that scores noisy diffusion latents directly, at small compute cost. Drop the resulting value model into any diffusion-alignment recipe (DPS, FK steering, DRaFT, DiffusionNFT, …) and each gets cheaper and often better at the same time.

See the project page for the full story, figures, and results.

---

🚧 Code release

The code release is in preparation. Star or watch this repository to be notified.

Citation

@article{go2026stitchvm,
    title   = {Stitched Value Model for Diffusion Alignment},
    author  = {Go, Hyojun and Chung, Hyungjin and Truong, Prune and Bhat, Goutam
               and Mi, Li and An, Zhaochong and Zhao, Zixiang and Narnhofer, Dominik
               and Belongie, Serge and Tombari, Federico and Schindler, Konrad},
    journal = {arXiv preprint arXiv:2605.19804},
    year    = {2026}
}

Contact

For questions about the method or paper, please open a GitHub issue or contact Hyojun Go.