Program synthesis by a language model is certainly cool, but I don't think it's really novel. After all, the mutator could as well just be a random AST transformer.
So what's the real deal here? Does the mutator end up with a usable program much quicker? Or does it cheat by looking at the test cases?
As an example, if I specify a md5 hash function on strings, what will that tool produce?
So what's the real deal here? Does the mutator end up with a usable program much quicker? Or does it cheat by looking at the test cases?
As an example, if I specify a md5 hash function on strings, what will that tool produce?