{
 "benchmark": "mathgate_exact_arithmetic",
 "generated_by": "scripts/run_mathgate_benchmark.py",
 "cases": 20000,
 "seed": 20260910,
 "verdicts": {
  "ALLOW": 11068,
  "BLOCK": 5945,
  "ABSTAIN": 2987
 },
 "wrong_claims": 5945,
 "wrong_claims_blocked": 5945,
 "wrong_claims_missed": 0,
 "correct_claims": 11068,
 "correct_claims_refused": 0,
 "binary_float_cases": 3940,
 "binary_float_errors_caught": 1402,
 "abstained_outside_grammar": 2987,
 "status": "MEASURED",
 "claim_boundary": "The gate checks that a claimed result equals the expression it is claimed for. It does not check that the expression is the right expression, and an exactly-correct answer to the wrong question passes. Anything outside its four-operator grammar is abstained rather than passed, and that count is reported rather than folded into the refusals. The expression is parsed, never executed. Synthetic cases."
}