We introduce clever, the first curated benchmark for evaluating the generation of specifications and formally verified code in lean The benchmark comprises of 161 programming problems One common approach is training models to refuse unsafe queries, but this strategy can be vulnerable to clever prompts, often referred to as jailbreak attacks, which can trick the ai into providing harmful responses Our method, stair (safety alignment with introspective reasoning), guides models to think more carefully before responding. Our analysis yields a novel robustness metric called clever, which is short for cross lipschitz extreme value for network robustness While, as we mentioned earlier, there can be thorny “clever hans” issues about humans prompting llms, an automated verifier mechanically backprompting the llm doesn’t suffer from these
A fundamental limitation of current ai agents is their inability to learn complex skills on the fly at test time, often behaving like “clever but clueless interns” in novel environments This severely limits their practical utility Hook it up with taskconfig—our handy layer for crafting clever input templates and grabbing outputs steadily via jmespath—and switching agents turns effortless, no extra fiddling needed Our benchmark structure ensures reproducibility by locking in versions.
WATCH
