Tofurther enhance thelearning ofthe twoalignment modules, an auxiliary loss is designed to contrast the foreground andbackgroundpixels.

In contrast to this, our algorithm works in the composite case with general maximally monotone operatorB(u).

This approach allows for formal subdifferentiation: forinstance, replacing derivativesbyClarkeJacobians in the usual differentiation formulas is fully justified for a wide class of nonsmooth problems.

Further,inthemiddlestage,i.e. fromthe7thto9thlayers,head tokens draw more attention from other tokens.