Introduction Current obfuscation research lacks of a large dataset that contains various realistic binaries (not only code snippets such as sorting algorithms), multiple obfuscators (not only OLLVM) ...