2 papers
cs.CV2026
K-MaT: Knowledge-Anchored Manifold Transport for Cross-Modal Prompt Learning in Medical Imaging
Jiajun Zeng, Shadi Albarqouni
Large-scale biomedical vision-language models (VLMs) adapted on high-end imaging (e.g., CT) often fail to transfer to frontline low-end modalities (e.g., radiography), collapsing i…
cs.CV2025
INTERACT-CMIL: Multi-Task Shared Learning and Inter-Task Consistency for Conjunctival Melanocytic Intraepithelial Lesion Grading
Mert Ikinci, Luna Toma, Karin U. Loeffler +6
Accurate grading of Conjunctival Melanocytic Intraepithelial Lesions (CMIL) is essential for treatment and melanoma prediction but remains difficult due to subtle morphological cue…