Preprint
Oct 2026
Reliability-Guided Vision-Language Decoding with Text-Aware Skip Refinement for Medical Image Segmentation
RELiF is a vision--language framework that integrates text through reliability-guided decoding that shows competitive performance across diverse imaging characteristics and target appearances and shows an association with prediction difficulty.
Sivan Talaei, Fatemeh Daneshfar, A. A. Abdullah et al.
· 0 citations