[WithinReason]

Adversarial networks are a straightforward solution to this. The reward for generating and solving tests is different.

[imtringued]

That's a good point. A model that is capable of implementing a nonsense test is still better than a model that can't. The implementer model only needs a good variety of tests. They don't actually have to translate a prompt into a test.