Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | https://arxiv.org/abs/2603.26790 |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866917364714438656 |
|---|---|
| author | Jones, Charles Noutahi, Emmanuel Hartford, Jason Eastwood, Cian |
| author_facet | Jones, Charles Noutahi, Emmanuel Hartford, Jason Eastwood, Cian |
| contents | Flow-matching generative models are increasingly used to simulate cell responses to biological perturbations. However, the design space for building such models is large and underexplored. We systematically analyse the design space of flow matching models for cell-microscopy images, finding that many popular techniques are unnecessary and can even hurt performance. We develop a simple, stable, and scalable recipe which we use to train our foundation model. We scale our model to two orders of magnitude larger than prior methods, achieving a two-fold FID and ten-fold KID improvement over prior methods. We then fine-tune our model with pre-trained molecular embeddings to achieve state-of-the-art performance simulating responses to unseen molecules.
Code is available at https://github.com/valence-labs/microscopy-flow-matching |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_26790 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Elucidating the Design Space of Flow Matching for Cellular Microscopy Jones, Charles Noutahi, Emmanuel Hartford, Jason Eastwood, Cian Computer Vision and Pattern Recognition Flow-matching generative models are increasingly used to simulate cell responses to biological perturbations. However, the design space for building such models is large and underexplored. We systematically analyse the design space of flow matching models for cell-microscopy images, finding that many popular techniques are unnecessary and can even hurt performance. We develop a simple, stable, and scalable recipe which we use to train our foundation model. We scale our model to two orders of magnitude larger than prior methods, achieving a two-fold FID and ten-fold KID improvement over prior methods. We then fine-tune our model with pre-trained molecular embeddings to achieve state-of-the-art performance simulating responses to unseen molecules. Code is available at https://github.com/valence-labs/microscopy-flow-matching |
| title | Elucidating the Design Space of Flow Matching for Cellular Microscopy |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2603.26790 |