Flow Matching with Uncertainty Quantification and Guidance
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866916006119604224 |
|---|---|
| author | Han, Juyeop Beyer, Lukas Lao Karaman, Sertac |
| author_facet | Han, Juyeop Beyer, Lukas Lao Karaman, Sertac |
| contents | Despite the remarkable success of sampling-based generative models such as flow matching, they can still produce samples of inconsistent or degraded quality. To assess sample reliability and generate higher-quality outputs, we propose uncertainty-aware flow matching (UA-Flow), a lightweight extension of flow matching that predicts the velocity field together with heteroscedastic uncertainty. UA-Flow estimates per-sample uncertainty by propagating velocity uncertainty through the flow dynamics. These uncertainty estimates act as a reliability signal for individual samples, and we further use them to steer generation via uncertainty-aware classifier guidance and classifier-free guidance. Experiments on image generation show that UA-Flow produces uncertainty signals more highly correlated with sample fidelity than baseline methods, and that uncertainty-guided sampling further improves generation quality. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_10326 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Flow Matching with Uncertainty Quantification and Guidance Han, Juyeop Beyer, Lukas Lao Karaman, Sertac Computer Vision and Pattern Recognition Machine Learning Despite the remarkable success of sampling-based generative models such as flow matching, they can still produce samples of inconsistent or degraded quality. To assess sample reliability and generate higher-quality outputs, we propose uncertainty-aware flow matching (UA-Flow), a lightweight extension of flow matching that predicts the velocity field together with heteroscedastic uncertainty. UA-Flow estimates per-sample uncertainty by propagating velocity uncertainty through the flow dynamics. These uncertainty estimates act as a reliability signal for individual samples, and we further use them to steer generation via uncertainty-aware classifier guidance and classifier-free guidance. Experiments on image generation show that UA-Flow produces uncertainty signals more highly correlated with sample fidelity than baseline methods, and that uncertainty-guided sampling further improves generation quality. |
| title | Flow Matching with Uncertainty Quantification and Guidance |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2602.10326 |