metadata
title: Formal Problem-Solving Exact Audit
emoji: 🧩
colorFrom: blue
colorTo: purple
sdk: static
pinned: false
tags:
- icml2026-repro
- paper-hgMZraPlSv
Beyond Theorem Proving: Formal Problem-Solving
Independent exact audit of all six registered claims using exhaustive logical semantics, all 1,086 released benchmark records and generated Lean files, and all 6,516 initial-release baseline method–problem slots.
Verdict: six verified claims; conservative maximum 12/12.