Crypto Accounting Bench: Evaluating Frontier and Open-Weight Models on Crypto-Asset Accounting Tasks
Deterministic diagnostics, read from each task's best of 3 attempts and macro-averaged across the 12 models, show higher base-amount agreement than deciding-account accuracy, which identifies account selection and complete-entry composition as the main remaining challenges on CAB.
Kareem A. Khattab, Omar Khattab, Mohamed I. Ibrahem
· 0 citations