How to finetune DINOv2 for astronmy?
This study evaluates the performance of existing visual foundation models, based on ViT, SwinV2, BEiT or DINOv2, for astronomical applications, particularly galaxy morphological …
About me
PhD student at IRISA, Université Bretagne Sud (expected 2026), and lecturer in computer science,
working on foundation models for astronomy and astrophysics, with a broader interest
in deep learning applied to scientific data.
Also contributing to the UniverseTBD collaboration on vision–language models for astronomy.
Interests
Foundation Models
Deep Learning
Astronomy & Astrophysics
Computer Vision
🔬 My Research
My work focuses on foundation models for astronomy and astrophysics, with applications such as galaxy morphological classification and contributions to observational programs involving transient events and aurorae.
I also contribute to the UniverseTBD collaboration, notably on AstroLLaVA, a vision–language model for astronomy. I am a member of the STAROS Projects team, which develops projects centered on spectroscopy for astronomy.
A complete list of my publications is available on ORCID and Google Scholar.
Featured Publications
AstroLLaVA: towards the unification of astronomical data and natural language
We present AstroLLaVA, a vision language model for astronomy that enables interaction with astronomical imagery through natural dialogue. By fine-tuning the LLaVA model on a …
sharaf-zaman
Recent Publications
(2025).
Orbital parameters of the bright binary alpha Draconis based on amateur spectra from the STAROS database.
arXiv preprint.
(2025).
How to finetune DINOv2 for astronmy?.
GRETSI 2025.
(2025).
AstroLLaVA: towards the unification of astronomical data and natural language.
ICLR 2025 Workshop.
Talks & Posters
Comment spécialiser DINOv2 pour l'astronomie?
Talk on evaluating and fine-tuning DINOv2 and other visual foundation models for galaxy morphological classification.
When Foundation Models Meet Astronomical Data
Poster on evaluating visual foundation models (ViT, DINOv2, CLIP) for astronomical tasks such as galaxy morphological classification.
When Foundation Models Meet Astronomical Data
ePoster — Session SS10: The impact of the rapidly evolving field of artificial intelligence on astrophysics research.