Auto-sync: 2026-06-30 04:35:29

results/paper_analysis.json

@@ -1,6 +1,6 @@
 {
   "best_clean_key": "residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmchallenger001",
-  "generated_utc": "2026-06-30T08:10:58+00:00",
+  "generated_utc": "2026-06-30T08:40:51+00:00",
   "mechanism_gap": {
     "best_clean_vs_direct_same_ckpt": 0.0776811594202898,
     "best_clean_vs_h16": 0.06318840579710144,
@@ -1432,6 +1432,103 @@
       "source": "results/h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_nmbonus0p02_summary.json",
       "std_success": 0.009889114266221456
     },
+    "residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmchallenger0005": {
+      "ci95_success": 0.03536553850393906,
+      "label": "K4 compatible tangents, no-op bonus 0.03, near-miss challenger gate 0.005",
+      "mean_action_mse_to_best": 0.4072380839080374,
+      "mean_progress": 0.5734310006964413,
+      "mean_success": 0.35999999999999993,
+      "num_completed": 3,
+      "per_task_success": {
+        "LiftPegUpright-v1": {
+          "mean_num_groups": 102.0,
+          "mean_success": 0.26340865087329823,
+          "std_success": 0.06057881385012102
+        },
+        "PickCube-v1": {
+          "mean_num_groups": 196.66666666666666,
+          "mean_success": 0.3292913809218157,
+          "std_success": 0.03124258859556991
+        },
+        "PullCube-v1": {
+          "mean_num_groups": 81.0,
+          "mean_success": 0.22889673679147363,
+          "std_success": 0.029397613100483843
+        },
+        "PushCube-v1": {
+          "mean_num_groups": 102.0,
+          "mean_success": 0.7571578319419062,
+          "std_success": 0.0856403530080274
+        },
+        "StackCube-v1": {
+          "mean_num_groups": 93.33333333333333,
+          "mean_success": 0.20044030044030045,
+          "std_success": 0.08896982358351117
+        }
+      },
+      "seed_action_mse_to_best": {
+        "0": 0.39503803297391404,
+        "1": 0.3989635453648541,
+        "2": 0.42771267338534413
+      },
+      "seed_progress": {
+        "0": 0.5589558385697234,
+        "1": 0.5713966150581836,
+        "2": 0.5899405484614165
+      },
+      "seed_success": {
+        "0": 0.3565217391304348,
+        "1": 0.34782608695652173,
+        "2": 0.37565217391304345
+      },
+      "selected_candidate_type_counts": {
+        "retrieval_residual_policy_residual": 1499,
+        "retrieval_residual_residual_near_miss": 122,
+        "retrieval_residual_residual_near_miss+residual_wrong_gripper": 5,
+        "retrieval_residual_residual_no_op": 56,
+        "retrieval_residual_residual_no_op+residual_wrong_gripper": 24,
+        "retrieval_residual_residual_wrong_gripper": 19
+      },
+      "selected_residual_scale_counts": {
+        "0.35": 1551,
+        "0.4": 21,
+        "0.45": 153
+      },
+      "selected_type_outcomes": {
+        "retrieval_residual_policy_residual": {
+          "count": 1499.0,
+          "mean_progress": 0.5646938670566709,
+          "success_rate": 0.34889926617745165
+        },
+        "retrieval_residual_residual_near_miss": {
+          "count": 122.0,
+          "mean_progress": 0.5747186182005728,
+          "success_rate": 0.38524590163934425
+        },
+        "retrieval_residual_residual_near_miss+residual_wrong_gripper": {
+          "count": 5.0,
+          "mean_progress": 0.7557869911193847,
+          "success_rate": 0.6
+        },
+        "retrieval_residual_residual_no_op": {
+          "count": 56.0,
+          "mean_progress": 0.7249722566921264,
+          "success_rate": 0.5
+        },
+        "retrieval_residual_residual_no_op+residual_wrong_gripper": {
+          "count": 24.0,
+          "mean_progress": 0.6711457151298722,
+          "success_rate": 0.4583333333333333
+        },
+        "retrieval_residual_residual_wrong_gripper": {
+          "count": 19.0,
+          "mean_progress": 0.6364115562878156,
+          "success_rate": 0.47368421052631576
+        }
+      },
+      "source": "results/h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_nmchallenger0p005_summary.json",
+      "std_success": 0.014235396124995555
+    },
     "residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmchallenger001": {
       "ci95_success": 0.034922898550724606,
       "label": "K4 compatible tangents, no-op bonus 0.03, near-miss challenger gate 0.01",
@@ -1529,6 +1626,103 @@
       "source": "results/h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_nmchallenger0p01_summary.json",
       "std_success": 0.014057223945486528
     },
+    "residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmchallenger0015": {
+      "ci95_success": 0.03115620450679224,
+      "label": "K4 compatible tangents, no-op bonus 0.03, near-miss challenger gate 0.015",
+      "mean_action_mse_to_best": 0.407323780797044,
+      "mean_progress": 0.5738324999127645,
+      "mean_success": 0.36000000000000004,
+      "num_completed": 3,
+      "per_task_success": {
+        "LiftPegUpright-v1": {
+          "mean_num_groups": 102.0,
+          "mean_success": 0.26340865087329823,
+          "std_success": 0.06057881385012102
+        },
+        "PickCube-v1": {
+          "mean_num_groups": 196.66666666666666,
+          "mean_success": 0.3294194734412126,
+          "std_success": 0.02715752817589447
+        },
+        "PullCube-v1": {
+          "mean_num_groups": 81.0,
+          "mean_success": 0.22889673679147363,
+          "std_success": 0.029397613100483843
+        },
+        "PushCube-v1": {
+          "mean_num_groups": 102.0,
+          "mean_success": 0.7571578319419062,
+          "std_success": 0.0856403530080274
+        },
+        "StackCube-v1": {
+          "mean_num_groups": 93.33333333333333,
+          "mean_success": 0.20044030044030045,
+          "std_success": 0.08896982358351117
+        }
+      },
+      "seed_action_mse_to_best": {
+        "0": 0.3947965855459156,
+        "1": 0.3992642119769817,
+        "2": 0.4279105448682347
+      },
+      "seed_progress": {
+        "0": 0.559589599908289,
+        "1": 0.5728900156656037,
+        "2": 0.5890178841644007
+      },
+      "seed_success": {
+        "0": 0.3565217391304348,
+        "1": 0.34956521739130436,
+        "2": 0.3739130434782609
+      },
+      "selected_candidate_type_counts": {
+        "retrieval_residual_policy_residual": 1544,
+        "retrieval_residual_residual_near_miss": 77,
+        "retrieval_residual_residual_near_miss+residual_wrong_gripper": 5,
+        "retrieval_residual_residual_no_op": 56,
+        "retrieval_residual_residual_no_op+residual_wrong_gripper": 24,
+        "retrieval_residual_residual_wrong_gripper": 19
+      },
+      "selected_residual_scale_counts": {
+        "0.35": 1575,
+        "0.4": 16,
+        "0.45": 134
+      },
+      "selected_type_outcomes": {
+        "retrieval_residual_policy_residual": {
+          "count": 1544.0,
+          "mean_progress": 0.5666152759130327,
+          "success_rate": 0.34974093264248707
+        },
+        "retrieval_residual_residual_near_miss": {
+          "count": 77.0,
+          "mean_progress": 0.5510438737253864,
+          "success_rate": 0.38961038961038963
+        },
+        "retrieval_residual_residual_near_miss+residual_wrong_gripper": {
+          "count": 5.0,
+          "mean_progress": 0.7557869911193847,
+          "success_rate": 0.6
+        },
+        "retrieval_residual_residual_no_op": {
+          "count": 56.0,
+          "mean_progress": 0.7249722566921264,
+          "success_rate": 0.5
+        },
+        "retrieval_residual_residual_no_op+residual_wrong_gripper": {
+          "count": 24.0,
+          "mean_progress": 0.6711457151298722,
+          "success_rate": 0.4583333333333333
+        },
+        "retrieval_residual_residual_wrong_gripper": {
+          "count": 19.0,
+          "mean_progress": 0.6364115562878156,
+          "success_rate": 0.47368421052631576
+        }
+      },
+      "source": "results/h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_nmchallenger0p015_summary.json",
+      "std_success": 0.012541047914657353
+    },
     "residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmchallenger002": {
       "ci95_success": 0.03115620450679224,
       "label": "K4 compatible tangents, no-op bonus 0.03, near-miss challenger gate 0.02",

results/paper_analysis.md

@@ -1,6 +1,6 @@
 # Paper Analysis
 
-Generated: `2026-06-30T08:10:58+00:00`
+Generated: `2026-06-30T08:40:51+00:00`
 
 ## Main Seed Statistics
 
@@ -54,6 +54,8 @@ Generated: `2026-06-30T08:10:58+00:00`
 | residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_oraclek8trace | K4 compatible tangents, no-op bonus 0.03, unique candidate-oracle prefix K=8 branch trace | 3 | 35.65% +/- 1.22 | +/- 3.02 | 57.16% | 0.406 | +5.91 pp |
 | residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmchallenger002 | K4 compatible tangents, no-op bonus 0.03, near-miss challenger gate 0.02 | 3 | 36.00% +/- 1.25 | +/- 3.12 | 57.42% | 0.407 | +6.26 pp |
 | residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmchallenger001 | K4 compatible tangents, no-op bonus 0.03, near-miss challenger gate 0.01 | 3 | 36.06% +/- 1.41 | +/- 3.49 | 57.38% | 0.407 | +6.32 pp |
+| residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmchallenger0005 | K4 compatible tangents, no-op bonus 0.03, near-miss challenger gate 0.005 | 3 | 36.00% +/- 1.42 | +/- 3.54 | 57.34% | 0.407 | +6.26 pp |
+| residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmchallenger0015 | K4 compatible tangents, no-op bonus 0.03, near-miss challenger gate 0.015 | 3 | 36.00% +/- 1.25 | +/- 3.12 | 57.38% | 0.407 | +6.26 pp |
 | residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmchallenger003 | K4 compatible tangents, no-op bonus 0.03, near-miss challenger gate 0.03 | 3 | 35.94% +/- 1.28 | +/- 3.18 | 57.36% | 0.407 | +6.20 pp |
 | residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmwgchallenger001 | K4 compatible tangents, no-op bonus 0.03, near-miss/wrong-gripper challenger gate 0.01 | 3 | 35.94% +/- 1.13 | +/- 2.81 | 57.40% | 0.424 | +6.20 pp |
 | residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmbonus001 | K4 composed compatible tangents, no-op bonus 0.03, singleton near-miss bonus 0.01 | 3 | 35.59% +/- 0.99 | +/- 2.46 | 57.10% | 0.406 | +5.86 pp |

results/paper_core_results.md

@@ -6,9 +6,9 @@ baseline is the h=16 rank-checkpoint online rollout (`29.74%`).
 For paired seed deltas, per-task gaps, and selection histograms, regenerate and
 read `paper_analysis.md` with `python3 scripts/build_paper_analysis.py`. Current
 paired analysis: best clean K4 compatible tangent transport with a trace-motivated
-near-miss challenger gate is `+6.26 pp` over canonical h=16, same-state no-expert
+near-miss challenger gate is `+6.32 pp` over canonical h=16, same-state no-expert
 lattice is `+27.25 pp`, and the remaining clean-to-same-state proposal gap is
-`+20.99 pp`.
+`+20.93 pp`.
 
 | Method | Uses same-state proposals | Uses expert proposal | Success | Gain vs policy | Interpretation |
 |---|---:|---:|---:|---:|---|
@@ -59,8 +59,9 @@ lattice is `+27.25 pp`, and the remaining clean-to-same-state proposal gap is
 | K4 composed type-consensus residual retrieval, no-op prior + exact near-miss/no-op incompatibility mask | No | No | 35.59% | +5.86 pp | Previous clean best; dropping only the weak `near_miss+no_op` composite preserves useful singleton/composite tangents and lowers action MSE |
 | K4 composed type-consensus residual retrieval, exact compatibility mask + no-op/near-miss singleton priors | No | No | 35.59% | +5.86 pp | Exact singleton near-miss priors 0.01/0.02 tie the top row and raise progress slightly; transferred near-miss tangents are high-precision but too sparse to close the proposal gap |
 | K4 compatible residual retrieval, unique candidate-oracle prefix K=8 | No | No | 43.07% diagnostic | +13.33 pp diagnostic | Diagnostic-only measured oracle over generated clean candidate prefix; unique-action trace shows real selector headroom, not duplicate-action inflation |
-| K4 compatible residual retrieval + near-miss challenger gate 0.02 | No | No | 36.00% | +6.26 pp | Current best clean deployment row; a two-stage selector keeps the robust compatible-chart anchor and only lets singleton near-miss tangents override when the field margin is positive |
-| K4 compatible residual retrieval, margin 0.10 | No | No | 34.67% | +4.93 pp | Naively lowering abstention selects too many bad nonzero tangents and falls below the 36.00% top row |
+| K4 compatible residual retrieval + near-miss challenger gate 0.01 | No | No | 36.06% | +6.32 pp | Current best clean deployment row; a two-stage selector keeps the robust compatible-chart anchor and only lets singleton near-miss tangents override under a tightly calibrated positive field margin |
+| K4 compatible residual retrieval + near-miss challenger fine calibration | No | No | 36.00-36.00% | +6.26 pp | Margins 0.005/0.015/0.02 form a near-tie plateau around the 0.01 optimum; 0.03 and near-miss+wrong-gripper are lower or higher-MSE, so the effect is calibrated singleton near-miss geometry rather than broad residual mixing |
+| K4 compatible residual retrieval, margin 0.10 | No | No | 34.67% | +4.93 pp | Naively lowering abstention selects too many bad nonzero tangents and falls below the 36.06% top row |
 | K4 composed type-consensus residual retrieval, exact compatibility mask + source-score prior | No | No | 35.48% | +5.74 pp | Measured train-source reward confidence does not replace the sparse typed prior on the compatible chart; useful negative calibration |
 | K4 composed type-consensus residual retrieval, exact compatibility mask + no-op/source-score priors | No | No | 35.54% | +5.80 pp | Adding measured train-source reward confidence to the current typed-prior chart ties near-best but does not beat exact compatibility masking alone |
 | K4 composed type-consensus residual retrieval, masked + component-wise no-op prior | No | No | 35.36% | +5.62 pp | Propagating the no-op prior into every composite over-selects composed tangents and drops below the exact-prior row; useful negative calibration |
@@ -113,7 +114,7 @@ Suggested main-table rows:
 14. K4 mean-by-type residual retrieval + scale-grid no-op prior 0.03
 15. K4 masked composed type-consensus residual retrieval + no-op prior 0.03
 16. K4 masked composed type-consensus residual retrieval + exact near-miss/no-op compatibility mask, with and without typed no-op prior
-17. K4 compatible tangent near-miss challenger gate 0.02
+17. K4 compatible tangent near-miss challenger gate 0.01 and fine margin calibration
 18. K4 masked composed type-consensus residual retrieval + source-score prior on the exact compatibility chart
 19. K4 masked composed type-consensus residual retrieval + component-wise no-op prior diagnostic
 20. K4 masked composed type-consensus residual retrieval + composite trust penalty diagnostic

results/paper_story_memo.md

@@ -16,7 +16,7 @@ when queried on proposal geometry that matches those local counterfactuals.
 | Same-state local counterfactual proposals are the mechanism | near-miss-only lattice is 55.94%; removing expert+near_miss drops to 25.57% | Strongly supported |
 | Conservative same-state result is large | no-expert lattice is 56.99% vs 29.74% policy | Main result |
 | Full lattice gives upper result | full lattice is 69.33%, oracle is 86.78% | Strong but label expert proposal clearly |
-| Deployment-clean proposal is currently a bottleneck | best clean K4 compatible tangent transport with a near-miss challenger gate is 36.00%, far below 56.99% | Supported |
+| Deployment-clean proposal is currently a bottleneck | best clean K4 compatible tangent transport with a near-miss challenger gate is 36.06%, far below 56.99% | Supported |
 | Gradient-based field optimization does not solve the clean proposal gap | `field_optim` best observed result is 25.39% | Negative diagnostic |
 | A broader non-expert proposal target does not reduce the proposal gap | direct broad non-expert policy is 27.88%; with field scoring it is 26.49% | Negative diagnostic |
 | Counterfactual residuals transfer better than absolute retrieved actions | nearest residual retrieval is 32.12% vs absolute retrieval 28.93%; KNN4 residual drops to 29.91% | Supported as a clean bridge |
@@ -33,7 +33,7 @@ when queried on proposal geometry that matches those local counterfactuals.
 | Train-source score priors do not explain the compatible tangent-chart gain | replacing the typed no-op prior with a measured train-source reward-score prior on the exact compatibility chart reaches 35.48%; adding it to the typed no-op row reaches 35.54%, still below the previous 35.59% compatibility row | Negative diagnostic: compatibility masking and sparse typed abstention matter more than continuous train-source confidence on this chart |
 | Singleton near-miss priors do not open a new clean route | adding a small exact `residual_near_miss` singleton prior 0.01 or 0.02 on the compatible chart ties the previous 35.59% compatibility row and slightly raises progress to 57.10% | Tie diagnostic: revived near-miss tangents are high-precision but too sparse to close the clean-to-same-state proposal gap |
 | Candidate-prefix oracle needs unique-action hygiene | the first K=8 prefix diagnostic was archived as `_nonunique`; the deduplicated unique-action trace reaches 43.07% candidate-oracle success with mean best branch rank 2.85 | Supported diagnostic: proposal headroom is real, but branch success falls with rank, so selector calibration must be conditional |
-| Near-miss challenger calibration improves clean deployment | the trace-motivated two-stage selector keeps the compatible-chart anchor and lets singleton near-miss tangents override only at margin 0.02; it reaches 36.00% and 57.42% progress | Current best clean row; small but story-aligned selector-calibration gain |
+| Near-miss challenger calibration improves clean deployment | the trace-motivated two-stage selector keeps the compatible-chart anchor and lets singleton near-miss tangents override only under a tightly calibrated margin; margin 0.01 reaches 36.06%, while 0.005/0.015/0.02 tie at 36.00% | Current best clean row; small but story-aligned selector-calibration gain, with wrong-gripper challenger variants rejected by lower success and higher MSE |
 | Component-wise composite priors do not add the gain | propagating the no-op bonus into composite types reaches 35.36%, below the exact-prior masked composition row at 35.54% and the exact compatibility row at 35.59% | Negative/near-tie diagnostic: sparse exact priors are better than broadly rewarding every no-op-containing composite |
 | Composite trust-radius penalty explains but does not improve the success top line | composite-only L2 penalty 0.02 ties 35.54% while lowering action MSE from 0.4106 to 0.4079; with the exact compatibility mask it lowers MSE further to 0.4048 but drops back to 35.54% | Tie/negative diagnostic: composed tangents need a local trust radius, but compatibility masking gives the success optimum |
 | Minimum-energy residual regularization does not add the gain | action L2 penalty 0.05 ties the previous 35.42% scale-grid row, while 0.10/0.20 reach 35.36% | Negative/tie diagnostic: the clean bridge is not explained by shortest-action bias |
@@ -47,7 +47,7 @@ when queried on proposal geometry that matches those local counterfactuals.
 | Typed no-op residual prior improves the clean bridge | CPU smoke `14883591` passed; bonuses 0.025/0.03/0.035 tie at 35.25%, while 0.01/0.02/0.05/0.08 are slightly lower | Fixed-scale clean diagnostic |
 | Wrong-gripper typed prior does not add a new clean bridge | wrong-gripper-only reaches 35.19%; no-op+wrong-gripper 0.02 ties 35.25%; no-op+wrong-gripper 0.04 drops to 35.13% | Negative/tie diagnostic |
 | No-op-only residuals nearly preserve the fixed-scale clean bridge | excluding wrong-gripper residuals gives 35.19% with either no-op bonus 0.03 or source-score bonus 0.02, one success below the 35.25% fixed-scale safe-family plateau | Mechanism sharpened: wrong-gripper is marginal, not core |
-| The proposal gap is now quantified | `paper_analysis.md` reports best clean +6.26 pp over canonical h16, same-state no-expert +27.25 pp, leaving a +20.99 pp clean-to-same-state gap | Core paper tension |
+| The proposal gap is now quantified | `paper_analysis.md` reports best clean +6.32 pp over canonical h16, same-state no-expert +27.25 pp, leaving a +20.93 pp clean-to-same-state gap | Core paper tension |
 | Policy fallback is not the same-state mechanism | adding a policy baseline candidate to the no-expert same-state lattice drops 56.99% to 40.70% even with margin 0.00 | Negative diagnostic |
 | Z-score retrieval metric does not help | z-score rows reach 32.23-32.81%, below raw retrieval | Negative diagnostic |
 | Task-relative actor-pose retrieval metric does not improve tangent transfer | K2 task-relative residual retrieval reaches 34.26% vs raw K2 35.01%; K4 task-relative mean-by-type + no-op reaches 34.43% vs raw K4 35.25% | Negative diagnostic |
@@ -83,7 +83,7 @@ clean proposal result, the intended main rows are:
 16. K4 mean-by-type tangent consensus + scale-grid typed no-op prior: 35.42%
 17. K4 masked composed type-consensus tangent transport: 35.30%; with exact near-miss/no-op compatibility mask only: 35.48%; with typed no-op prior: 35.54%; with exact compatibility mask + typed no-op prior: 35.59%; adding singleton near-miss prior 0.01/0.02 ties 35.59%; with source-score prior on that exact chart: 35.48%; with typed no-op + source-score: 35.54%
 18. K4 compatible tangent unique-action candidate-oracle prefix K=8 diagnostic: 43.07%; branch trace shows mean best branch rank 2.85, unique count 8.00, and real but conditional selector headroom
-19. K4 compatible tangent near-miss challenger gate 0.02: 36.00%; current best clean deployment row from trace-motivated two-stage selector calibration
+19. K4 compatible tangent near-miss challenger gate 0.01: 36.06%; current best clean deployment row from trace-motivated two-stage selector calibration. Fine margins 0.005/0.015/0.02 tie at 36.00%, while 0.03 and near-miss+wrong-gripper variants are lower or higher-MSE.
 20. K4 compatible tangent abstention margin 0.10 diagnostic: 34.67%; naive lower abstention hurts despite candidate-prefix oracle headroom
 20. K4 masked composed type-consensus tangent transport + component-wise no-op prior: 35.36%
 21. K4 masked composed type-consensus tangent transport + composite L2 trust penalty: 35.54% at 0.02; 35.48% at 0.05; 35.54% with exact compatibility mask + L2 0.02
@@ -148,7 +148,7 @@ masked composition reached 35.30%, exact compatibility masking without typed
 or source-score priors reached 35.48%, masked composition with the exact typed
 no-op prior reached 35.54%, and raw selected candidate types show no
 random-negative or wrong-direction composite leak. The paper table and paired
-analysis now use the exact compatibility-mask row as `best_clean_key`. The
+analysis now use the near-miss challenger margin `0.01` row as `best_clean_key`. The
 component-wise composite prior follow-up completed at 35.36%, below the exact
 no-op prior row. Composite-only L2 trust penalties reduce action MSE but do not
 improve success: standalone L2 0.02 ties 35.54%, L2 0.05 reaches 35.48%, and
@@ -168,9 +168,12 @@ and blocks invalid padding branches from winning. The unique-action rerun
 `14953513`/`14953522` completed at 43.07% candidate-oracle success, and the
 branch-trace rerun `14953960`/`14953961` confirmed unique count 8.00 and mean
 best branch rank 2.85. A trace-motivated clean near-miss challenger gate
-(`14954280`/`14954281`) completed at 36.00%, the current clean best. The clean
-compatible-chart margin `0.10` test completed at 34.67%, so simple lower
-abstention is not the answer.
+completed at 36.00% for margin `0.02` (`14954280`/`14954281`), then improved to
+36.06% for margin `0.01` (`14954526`/`14954527`), the current clean best. Fine
+calibration jobs for margins `0.005` and `0.015` (`14955089`/`14955090` and
+`14955093`/`14955094`) tie 36.00%, confirming a tight selector-calibration
+plateau around the 0.01 optimum. The clean compatible-chart margin `0.10` test
+completed at 34.67%, so simple lower abstention is not the answer.
 
 - `14858328`-`14858333`: completed train-split `field_selected_noexpert_bc5`;
   direct rollout is 26.84%, field-guided best is 27.65%.
@@ -432,7 +435,7 @@ abstention is not the answer.
   no-op prior, field-gated tangent length calibration over `0.35/0.40/0.45`, an
   exact near-miss/no-op compatibility mask, and the trace-motivated singleton
   near-miss challenger gate as the current best clean deployment diagnostic,
-  36.00%, not as a SOTA claim. The previous mean-by-type scale-grid no-op row
+  36.06%, not as a SOTA claim. The previous mean-by-type scale-grid no-op row
   remains 35.42%; pure masked composition is 35.30%, the exact typed-prior
   composition row is 35.54%, exact compatibility reaches 35.59%, and
   component-wise propagation of the no-op prior reaches 35.36%, so composition

results/paper_table_status.json

@@ -1429,6 +1429,44 @@
       "best_config": null,
       "gain_vs_h16_policy": 0.06318840579710144
     },
+    {
+      "key": "retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmchallenger0005",
+      "label": "K4 compatible residual retrieval, near-miss challenger gate 0.005",
+      "path": "h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_nmchallenger0p005_summary.json",
+      "clean_deployment": "yes",
+      "same_state_proposals": "no",
+      "expert_proposal": "no",
+      "story_role": "fine lower-margin calibration for the trace-motivated singleton near-miss challenger gate",
+      "fallback_success": null,
+      "pending_job": "14955089/14955090",
+      "path_exists": true,
+      "status": "complete",
+      "success": 0.36,
+      "std_success": 0.014235396124995555,
+      "completed_seeds": null,
+      "num_completed": 3,
+      "best_config": null,
+      "gain_vs_h16_policy": 0.06260869565217392
+    },
+    {
+      "key": "retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmchallenger0015",
+      "label": "K4 compatible residual retrieval, near-miss challenger gate 0.015",
+      "path": "h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_nmchallenger0p015_summary.json",
+      "clean_deployment": "yes",
+      "same_state_proposals": "no",
+      "expert_proposal": "no",
+      "story_role": "fine upper-margin calibration for the trace-motivated singleton near-miss challenger gate",
+      "fallback_success": null,
+      "pending_job": "14955093/14955094",
+      "path_exists": true,
+      "status": "complete",
+      "success": 0.36,
+      "std_success": 0.012541047914657353,
+      "completed_seeds": null,
+      "num_completed": 3,
+      "best_config": null,
+      "gain_vs_h16_policy": 0.06260869565217392
+    },
     {
       "key": "retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmchallenger003",
       "label": "K4 compatible residual retrieval, near-miss challenger gate 0.03",

results/paper_table_status.md

@@ -78,6 +78,8 @@ Baseline h=16 policy: 29.74%
 | retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_oraclek8trace | K4 compatible residual retrieval, unique candidate-oracle prefix K=8 branch trace | complete | 43.07% | +13.33 pp | diagnostic | no | no | diagnostic branch trace for deployable selector calibration over the local tangent chart |
 | retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmchallenger002 | K4 compatible residual retrieval, near-miss challenger gate 0.02 | complete | 36.00% | +6.26 pp | yes | no | no | trace-motivated two-stage selector calibration over the compatible local tangent chart |
 | retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmchallenger001 | K4 compatible residual retrieval, near-miss challenger gate 0.01 | complete | 36.06% | +6.32 pp | yes | no | no | lower-margin sensitivity for trace-motivated singleton near-miss challenger calibration |
+| retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmchallenger0005 | K4 compatible residual retrieval, near-miss challenger gate 0.005 | complete | 36.00% | +6.26 pp | yes | no | no | fine lower-margin calibration for the trace-motivated singleton near-miss challenger gate |
+| retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmchallenger0015 | K4 compatible residual retrieval, near-miss challenger gate 0.015 | complete | 36.00% | +6.26 pp | yes | no | no | fine upper-margin calibration for the trace-motivated singleton near-miss challenger gate |
 | retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmchallenger003 | K4 compatible residual retrieval, near-miss challenger gate 0.03 | complete | 35.94% | +6.20 pp | yes | no | no | upper-margin sensitivity for trace-motivated singleton near-miss challenger calibration |
 | retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmwgchallenger001 | K4 compatible residual retrieval, near-miss/wrong-gripper challenger gate 0.01 | complete | 35.94% | +6.20 pp | yes | no | no | trace-motivated challenger calibration test for whether wrong-gripper residuals carry conditional selector headroom |
 | retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmbonus001 | K4 composed compatible residual retrieval, no-op bonus 0.03, singleton near-miss bonus 0.01 | complete | 35.59% | +5.86 pp | yes | no | no | revive high-precision singleton near-miss tangents without boosting toxic near-miss+no-op composites |

scripts/build_paper_table_status.py

@@ -773,7 +773,7 @@ SPECS = [
         same_state_proposals="no",
         expert_proposal="no",
         story_role="fine lower-margin calibration for the trace-motivated singleton near-miss challenger gate",
-        pending_job="14954890/14954893",
+        pending_job="14955089/14955090",
     ),
     ResultSpec(
         key="retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmchallenger0015",
@@ -783,7 +783,7 @@ SPECS = [
         same_state_proposals="no",
         expert_proposal="no",
         story_role="fine upper-margin calibration for the trace-motivated singleton near-miss challenger gate",
-        pending_job="14954894/14954895",
+        pending_job="14955093/14955094",
     ),
     ResultSpec(
         key="retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmchallenger003",