SRWKV: A Shape-Guided RWKV With Adaptive Receptive Fields for Efficient Medical Image Segmentation.
SRWKV is proposed, a shape-guided RWKV (SGR) model for parameter-efficient MedISeg that introduces an SGR block that uses a shape prior predicted from the deepest encoder feature to guide token traversal during decoding, reducing foreground-background interleaving and improving structural coherence during sequence form...