Semi-formal reasoning: force the agent to state premises and trace execution before it verifies a patch

arXiv·low signal

Agentic Code Reasoning has the model construct explicit premises, trace concrete execution paths, and derive a formal conclusion instead of eyeballing correctness — pushing patch-equivalence accuracy from 78% to 88% on curated cases and 93% on real agent-generated patches. It's a structured-prompting upgrade over free-form chain-of-thought for any verification task where the answer must actually hold, not just look plausible.