04:00
2026-05-27
arxiv.org
machine-learning
Evi-Steer: Learning to Steer Biomedical Vision-Language Models through Efficient and Generalizable Evidential Tuning
Researchers have developed Evi-Steer, a new framework for BiomedCLIP that enables uncertainty-aware fine-tuning of biomedical vision-language models by updating only 0.11% of total parameters. The met…