Auto-sync: 2026-06-30 04:47:32

results/README.md

@@ -147,9 +147,12 @@ action MSE but do not improve success: standalone L2 `0.02` ties 35.54%, L2
 The unique-action K=8 candidate-oracle diagnostic is complete: the archived
 non-unique run remains `_nonunique`, while the deduplicated branch trace reaches
 43.07% candidate-oracle success with mean best branch rank 2.85. The resulting
-trace-motivated near-miss challenger gate (`residual_near_miss`, margin 0.02)
-is deployment-clean and reaches 36.00% mean success with 57.42% progress, the
-current clean best.
+trace-motivated near-miss challenger gate (`residual_near_miss`, margin 0.01)
+is deployment-clean and reaches 36.06% mean success with 57.38% progress, the
+current clean best. Fine margins `0.005`, `0.015`, and `0.02` tie at 36.00%,
+while `0.03` is slightly lower and adding wrong-gripper challengers raises MSE,
+so the clean gain is a tight singleton near-miss calibration rather than broad
+residual mixing.
 
 - Code supports `--retrieval-residual-scale`, defaulting to `1.0` so the
   completed 32.12% result is unchanged.
@@ -395,9 +398,12 @@ current clean best.
   Jobs `14953960`/`14953961` completed the branch-trace rerun used for selector
   calibration.
 - Jobs `14954280`/`14954281` completed the trace-motivated near-miss challenger
-  gate clean deployment run at 36.00%, the current clean best.
+  gate clean deployment run at margin `0.02` with 36.00%; jobs
+  `14954526`/`14954527` completed margin `0.01` at 36.06%, the current clean
+  best; jobs `14955089`/`14955090` and `14955093`/`14955094` completed fine
+  margin checks at `0.005` and `0.015`, both 36.00%.
 - Jobs `14936131`/`14936132`/`14936133` completed the compatible-chart
-  abstention test at margin `0.10`: it reaches 34.67%, below the 36.00% top row.
+  abstention test at margin `0.10`: it reaches 34.67%, below the 36.06% top row.
   This rules out a naive lower-margin fix for the apparent oracle headroom.
 
 Paper table automation: