The paper outlines a prototype or method for pruning infeasible paths in path-biased random testing which is helpful in eliminating useless and unexploitable drawings.
It utilizes graph transformation to preserve the actual behavior of the program.
The prototype improves the control flow graph of a program and builds a transformed graph with fewer infeasible paths.
Further, the prototype is driven by symbolic execution and constraint solving that use detection of subsumptions and the abstract-check-refine paradigm to determine the infeasibility of some paths including cyclic paths.
What do you think?