Year: 2026
Type: preprint
Abstract: We investigate whether specially constructed text captions can capture the same morphological information as radio galaxy images. Using the MiraBest dataset, we generate captions with a domain-specific prompt and evaluate their alignment with images through the SigLIP-2 vision--language model, with and without LoRA fine-tuning. Results show that caption-based classification of FR-I and FR-II galaxies performs similarly to images, with fine-tuning improving local coherence of embeddings but not global alignment.
Source: arXiv (Cornell University)
Institution: University of Geneva
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7167706289
Cites:
Cited by:
Related to: