Add fail-closed failure-driven YOLO sampling

docs/BELGIUM_BUILDING_TRAINING_LOOP.md

@@ -79,6 +79,14 @@ The active production model remains unchanged while any gate fails.
 
 Every failed assessment returns `continue_training_loop`. Only a report with
 `training_complete` may proceed to final human review and guarded activation.
+
+After a failed assessment,
+`scripts/build_failure_driven_yolo_sampling.py` creates a checksummed,
+train-only sampling manifest. Positive tiles from regions that fail F1 or
+recall are repeated, while true negative train tiles are repeated when a
+regional precision gate or the pure-background gate fails. Calibration, test,
+background-test and validation AOIs are excluded by their frozen corpus split;
+the generated evidence records that no protected sample entered training.
 The orchestrator refuses to start unless the frozen dataset audit is `ok` and
 contains zero blank/low-variance positive tiles.
 For dated imagery, GRB `BEGINDATUM` and PICC `DATE_CREAT` are compared with the