AI RESEARCH
Towards Responsible Multimodal Medical Reasoning via Context-Aligned Vision-Language Models
arXiv CS.CV
•
ArXi:2604.08815v1 Announce Type: new Medical vision-language models (VLMs) show strong performance on radiology tasks but often produce fluent yet weakly grounded