nx_pose_benchmark_audit.nx
buildroot/runtime/nx_pose_benchmark_audit.nx
about
nx_pose_benchmark_audit.nx -- CRITIC + ADVERSARY audit (operator 2026-07-14: "make sure we aren't self validating
but getting INDUSTRY benchmarking and grading... use the researcher, census, critic, adversary"). Grades every
pose/build-2 CLAIM against the GROUNDED industry standard (MS COCO Keypoint AP, OKS-based; ViTPose 75.8-81.1 AP,
researcher-fetched to knowledge/library/vitpose_modelcard_coco_ap.md), tagging each MECHANICAL (gate-verified
computation = legit) vs SELF-VALIDATED (circular/eyeball = the violation) vs INDUSTRY-MEASURED (external = what we
OWE). Each row carries the ADVERSARY's refutation. Liar-killed: cited evidence must open. VERDICT is deliberately
harsh -- ZERO industry benchmarks is the honest truth. license_tier: ORIGINAL expect_exit: 0
dependencies 1 imports · 0 importers
imports: nx_gate.nx
imported by: nobody (leaf or entry point)
call flow from main pre-order; caps 40 nodes / depth 6 declared; ↻ = already shown
structs
| none |
consts
| none |
functions
| 10 | func ab_open(path: *u8) -> i64 { let fd: i64 = sys_openat_rd(path); if fd < 0 { return 0 } sys_close(fd); return 1 } |
| 14 | func ab_claim(claim: *u8, grade: i64, evidence: *u8, adversary: *u8, c: *i64) -> i64 |
| 29 | func main() -> i64 |