Auto-sync: 2026-06-28 23:20:28

results/README.md

@@ -93,9 +93,10 @@ Completed deployment-clean diagnostics:
 
 Completed residual transport jobs:
 
-Last checked: `2026-06-29 02:45 UTC`. The task-relative retrieval-metric batch
-completed after passing CPU/unit smokes, and the paper table rebuild now includes
-the task-relative rows.
+Last checked: `2026-06-29 03:20 UTC`. The task-relative retrieval-metric batch
+completed after passing CPU/unit smokes, and the source-progress viability-gate
+jobs are queued behind completed smokes. The paper table marks the source-progress
+rows pending with the new job IDs.
 
 - Code supports `--retrieval-residual-scale`, defaulting to `1.0` so the
   completed 32.12% result is unchanged.
@@ -195,6 +196,14 @@ the task-relative rows.
   no-op 0.03 reaches 34.43%, and K2 safe residual retrieval reaches 34.26%;
   both are below the raw-metric rows at 35.25% and 35.01%. Summary jobs
   `14893788`/`14893790` and rebuild job `14893791` completed.
+- Job `14894281` completed the unit smoke for the source-progress viability gate,
+  including the variable residual-count padding check.
+- Job `14894282` completed the ManiSkill CPU rollout smoke for K4 mean-by-type
+  residual retrieval with `source progress >= 0.50`; it used
+  `GROUP_BATCH_SIZE=1` to respect the CPU backend's single-env constraint.
+- Jobs `14894298`/`14894299` are queued GPU arrays for source-progress thresholds
+  `0.50` and `0.75`, with summaries `14894300`/`14894301` and rebuild job
+  `14894302` dependent on completion.
 - Job `14869627` completed the CPU Apptainer smoke for the residual scale-grid
   selector: selected index `3` and returned action `0.20` on the toy ray-search
   case, validating candidate expansion before rollout.