audit nested YOLO labels without destructive rewrites

docs/CODEX_EXECUTION_LOG.md

@@ -12561,4 +12561,7 @@ Open:
 - A read-only same-class box probe found zero exact duplicates and zero pairs
   at IoU >= 0.90. It flagged 43 possible containment pairs across 36 tiles for
   human adjudication. No source label was modified or silently excluded.
-- Targeted renderer tests passed: 6 tests.
+- Added a reproducible relationship auditor and rendered the 36 flagged tiles
+  on a dedicated 384 px contact sheet. Visual follow-up found complex/adjacent
+  building components rather than a systematic duplicate-label pattern.
+- Targeted auditor and renderer tests passed: 9 tests.

docs/TODO.md

@@ -1155,6 +1155,8 @@ This file now starts with the current implementation status. Older preparation/b
 - [ ] Human-adjudicate the 43 possible nested same-class box pairs in 36 tiles;
   do not rewrite or exclude them automatically because GRB objects can overlap
   legitimately.
+- [x] Persist a deterministic label-relationship audit and a focused 36-tile
+  contact sheet; AI follow-up found no systematic duplicate-label pattern.
 - [ ] Convert the AI-assisted ledger into no stronger claim than experimental
   triage; a real human must independently review and sign the frozen artifacts
   before the governed training wrapper may unlock.

docs/reviews/2026-08-09-ai-assisted-checkpoint-review.md

@@ -111,3 +111,9 @@ or low-variance tiles. No exact or IoU>=0.90 duplicate box pair was found. A
 separate containment probe identified 43 potentially nested pairs across 36
 tiles (0.0543% relative to rendered labels); these remain human-adjudication
 candidates and were neither rewritten nor automatically excluded.
+
+The 36 flagged tiles were then rendered on a separate high-resolution sheet.
+AI-assisted inspection found no systematic duplicate-label pattern: the
+relationships predominantly represent adjacent or complex building components
+in dense GRB contexts. All tiles remain available for experimental analysis,
+while the exact 43 pairs stay visible for human release adjudication.