Auto-sync: 2026-06-28 20:44:54 (part 2)

results/README.md

@@ -91,7 +91,7 @@ Completed deployment-clean diagnostics:
 
 Completed residual transport jobs:
 
-Last checked: `2026-06-28 22:10 UTC`. The counterfactual tangent ray-search batch
+Last checked: `2026-06-29 00:41 UTC`. The counterfactual tangent ray-search batch
 completed; the typed no-op residual-prior clean sweep also completed after a
 passing CPU smoke.
 
@@ -158,6 +158,16 @@ passing CPU smoke.
   35.19%, 35.19%, 35.25%, and 35.25%, so `0.025`/`0.03`/`0.035` form the
   current best clean plateau. Summary jobs `14884376`/`14884378`/`14884380`/
   `14884382` and rebuild job `14884383` completed.
+- Job `14890019` completed the CPU smoke for multi-type candidate bonuses
+  (`residual_no_op=0.03`, `residual_wrong_gripper=0.02`), confirming the
+  CLI/Slurm path for multiple typed priors.
+- Jobs `14890071`/`14890073`/`14890075`/`14890077` completed wrong-gripper and
+  no-op+wrong-gripper typed-prior sweeps. Results are 35.19% for
+  wrong-gripper-only, 35.25% for no-op 0.03 + wrong-gripper 0.02, 35.13% for
+  no-op 0.03 + wrong-gripper 0.04, and 35.25% for no-op 0.025 + wrong-gripper
+  0.02. These are tie/negative diagnostics, not a new clean best. Summary jobs
+  `14890072`/`14890074`/`14890076`/`14890078` and rebuild job `14890079`
+  completed.
 - 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.