docs: record dynamic projection stability and short pass

3 files changed, 63 insertions, 4 deletions

diff --git a/RESULTS.md b/RESULTS.md
index 7c9ddd3..0a68133 100644
--- a/RESULTS.md
+++ b/RESULTS.md
@@ -961,6 +961,40 @@ freezes the predictor, restores loader state, and evaluates no validation or
 test example. Sign-only dissipativity is therefore closed before endpoint
 access; no margin is selected and the score remains 5/10.
 
+The new dynamic neutral-projection branch does not weaken those negative stop
+rules. Instead of another fixed margin, it uses a fast paired instruction-off
+soma/apical observation to remove only the current affine component of the
+*neutral* residual. The slow 64-example predictor remains frozen and the fast
+fit sees zero task-instruction observations. The frozen D1 training-only gate
+passes all 14 checks for 352 steps: maximum loss is `2.9281`, final-32 mean
+loss is `1.6293`, maximum forward/feedback weights are `1.181/0.833`, and all
+optimizer and BatchNorm states remain finite. The uncontrolled neutral
+residual grows to `0.005833` of traffic RMS with a correction slope as large
+as `0.02968`; projection holds the residual ratio and soma slope to
+`3.03e-8` and `1.52e-8`. Thus the controller is active rather than benefiting
+from an accidentally exact initialization. D1 evaluates neither validation
+nor test and cannot change the paper score.
+
+The subsequently frozen D2 short validation endpoint also passes every gate:
+
+| method | validation accuracy | early alignment | epoch-11--20 feedback cosine | estimated MACs |
+|:--|--:|--:|--:|--:|
+| dynamic neutral-projection innovation | **83.58%** | 0.883830 | 0.860594 | 1.4519e14 |
+| clean KP | 82.66% | 0.885583 | 0.854594 | 1.4519e14 |
+| frozen MT-1 raw / matched | 10.00% / 10.00% | nonfinite | nonfinite | 1.4530e14 each |
+
+D2 is finite for all 20 epochs and ends with validation loss `0.511322` and
+feedback/forward cosine `0.904056`. The unprojected neutral/traffic ratio
+reaches `0.009561`, while the projected remainder stays below `3.04e-8` with
+maximum soma slope `1.44e-8`; used-signal/instruction RMS differs by at most
+`4.67e-10`. It uses zero task-loss queries, `1.3261x` BP MACs, 4.070e12
+separately reported elementwise operations, one paired neutral observation per
+ordinary example, 2.165 GB peak allocated memory, one final validation
+evaluation, and no test evaluation. This is the first stable load-bearing
+innovation result on the standard ResNet substrate, but it is still a
+one-seed short validation record. Per the frozen scoring rule it leaves the
+reviewer score at 5/10 and opens only the predeclared D3 full validation run.
+
 ## How to run
 `experiments/run.py --mode {bp,fa,dfa,sdil} --dataset {mnist,fmnist,cifar10} --depth D --residual {0,1} --act {tanh,gelu,silu,relu}`
 Batteries: `experiments/run_v2.sh <ds> "<depths>" <res> <act> "<seeds>" <ep> <pfx>`.
diff --git a/REVIEW_SCORECARD.md b/REVIEW_SCORECARD.md
index 5443510..4257ac3 100644
--- a/REVIEW_SCORECARD.md
+++ b/REVIEW_SCORECARD.md
@@ -25,9 +25,9 @@ Every formal result report records:
 | Soundness | 3/4 | Theory, local-gradient checks, causal diagnostics, cost accounting, and frozen stop rules are unusually careful. The learned apical vectorizer remains an unresolved failure mode. |
 | Novelty | 2/4 | Learned node-perturbation feedback is prior art. The defensible novelty is the per-cell innovation operation under mixed apical traffic, together with its causal and scaling analysis. |
 | Significance | 3/4 | Near-flat performance over 12x depth while DFA alignment collapses is potentially important, but the current flattened-CIFAR task does not benefit from depth and the frozen standard-ResNet recipe failed. |
-| Empirical support | 2/4 | Five-depth scaling and the residualization ablation are strong. Useful-depth C2, broad endogenous C1, oral-B, full ResNet A3, and KP mixed-traffic MT-1 gates failed; untouched confirmation panels correctly remained sealed. |
+| Empirical support | 2/4 | Five-depth scaling and the residualization ablation are strong. Dynamic neutral projection now passes one standard-ResNet short validation gate, but useful-depth C2, broad endogenous C1, oral-B, full ResNet A3, and the original KP mixed-traffic MT-1 gates failed; full D3 and confirmation remain open. |
 | Reproducibility | 4/4 | Code, exact provenance, seed panels, costs, failed branches, frozen selectors, and staged test-access rules are retained in git. |
-| **Overall** | **5/10** | **Borderline reject: a strong core result without a completed standard-scale or useful-depth demonstration.** |
+| **Overall** | **5/10** | **Borderline reject: a strong core plus a promising short standard-scale result, but no completed full or independently confirmed endpoint.** |
 | Confidence | 4/5 | High confidence in the assessment because the positive and negative branches are both extensively audited. |
 
 ### Evidence already carrying the paper
@@ -48,8 +48,9 @@ Every formal result report records:
    supported mechanism is narrower than the initial claim.
 4. The frozen oral-B screen falsified the desired-velocity and Harnett error-derivative claims.
 5. The frozen ResNet-20 A3 run became nonfinite and ended at chance; the later KP mixed-traffic
-   MT-1 panel also became nonfinite under all three signals. A4, MT-2, and MT-3 remain sealed, so
-   completed native baselines improve fairness but do not supply SDIL scale evidence.
+   MT-1 panel also became nonfinite under all three signals. Dynamic neutral projection repairs a
+   20-epoch seed-0 validation run, but its full D3 endpoint and independent confirmation are not
+   complete. A4, MT-2, and MT-3 remain sealed.
 
 ## Score trajectory and prospective gates
 
@@ -73,6 +74,9 @@ Every formal result report records:
 | Mixed-traffic MT-1 fails | 5 | Raw, norm-matched raw, and innovation all become nonfinite in epoch 1 and end at 10%; calibration and predictor warmup still pass | Closes the controlled standard-ResNet innovation path; MT-2/MT-3 remain untouched and no weaker traffic rescue is allowed |
 | Mixed-traffic MT-2 | not opened | The prerequisite MT-1 gate failed | No full seed-0 validation claim is available |
 | Mixed-traffic MT-3 | not opened | MT-2 was never opened; all five confirmation seeds remain untouched | No controlled ResNet innovation confirmation claim is available |
+| Dynamic projection D1 | 5 | All 352 training-only steps remain finite while the fast neutral fit holds residual coupling near zero | Mechanics only; no held-out endpoint |
+| Dynamic projection D2 | 5 | One frozen 20-epoch record reaches 83.58%, above clean KP's 82.66%, with all stability and cost gates passing | Short single-seed validation cannot establish the full scaling claim |
+| Dynamic projection D3 | running | A predeclared 200-epoch near-BP validation gate is open | Only a complete pass can make 6/10 eligible and open confirmation |
 | Oral-A A4 | not opened | The prerequisite A3 gate failed | No oral-A confirmation claim is available |
 
 These are conditional reviewer forecasts, not promised scores. A failed stage leaves its negative
@@ -104,6 +108,8 @@ retroactively reopened by success on standard vision benchmarks.
 | 2026-07-22 / `f310ad5` mixed-traffic MT-1 | All three frozen conditions become nonfinite in epoch 1 and end at 10%; ratio calibration, predictor warmup, zero-query, and cost checks pass | 5 → 5 | Fails to make innovation load-bearing on a standard ResNet and closes MT-2/MT-3; the prior mechanism-only evidence survives but empirical support does not improve |
 | 2026-07-22 / `6531aed`, `d818fc2` MT-1 diagnosis | Training-only localization finds the first active failure at stem steps 23/69/74 for raw/matched/innovation; residual coupling yields the verified multiplicative operator `D W C` | 5 → 5 | Sharpens the failure into an operator-stability requirement and improves soundness of the negative analysis, but adds no successful held-out SDIL endpoint |
 | 2026-07-22 / `8d28fd9` stability S0 | A frozen training-only four-margin grid has no eligible candidate; sign-certified margins either overflow BN state or produce enormous transient loss and parameter growth | 5 → 5 | Closes fixed one-sided predictor bias and motivates a two-sided dynamic gain condition, but no validation endpoint or empirical support is gained |
+| 2026-07-22 / `6b85249` dynamic projection D1 | The sole frozen training-prefix record passes 14/14 checks; a 0.0058 neutral residual is reduced to 3.0e-8 while weights, momentum, BN, and loss stay bounded | 5 → 5 | Establishes a nontrivial operator-stability repair, but training-only mechanics cannot change the recommendation |
+| 2026-07-22 / `9bc58d6` dynamic projection D2 | The sole frozen short validation record reaches 83.58% versus clean KP's 82.66% and failed controls' 10%, with 0.884 early alignment, zero queries, and 1.326x BP MACs | 5 → 5 | First positive standard-ResNet load-bearing innovation endpoint, but one short seed is insufficient; full D3 and independent confirmation remain decisive |
 
 Future rows are appended only after an audited frozen stage. A score staying flat is informative:
 engineering, theory exposition, or visualization may make the paper more defensible without
diff --git a/ROADMAP.md b/ROADMAP.md
index 35b1ed4..57b6710 100644
--- a/ROADMAP.md
+++ b/ROADMAP.md
@@ -515,6 +515,25 @@ margins retain finite tensors but violate loss, signal-ratio, weight, and
 momentum envelopes by orders of magnitude. No validation protocol opens. This
 rules out fixed sign-only predictor bias as the missing gain-control mechanism.
 
+**Dynamic neutral projection D1: passed.** `DYNAMIC_INNOVATION.md` freezes a
+single fast local controller rather than another margin grid. Every task batch
+provides a paired instruction-off soma/apical observation; the fast affine fit
+removes the current neutral residual coupling without reading task instruction
+or changing the slow predictor. All 352 training-only steps and states remain
+finite, maximum loss is `2.9281`, and the controller reduces a growing
+`0.005833` neutral/traffic ratio to at most `3.03e-8`. No held-out evaluation
+occurs, so this mechanics pass opens D2 without changing the score.
+
+**Dynamic neutral projection D2: passed.** The sole frozen 20-epoch record
+reaches `83.58%` validation accuracy, above clean KP's `82.66%` and 73.58
+points above both failed MT-1 controls. Early alignment is `0.883830`, final
+feedback cosine is `0.904056`, all projection certificates pass, cost is
+`1.3261x` BP MACs plus explicitly reported elementwise work, and test remains
+untouched. Because this is one short seed, the score stays 5/10. The
+predeclared 200-epoch D3 validation endpoint is now open; only a complete D3
+pass can make the 5-to-6 score update eligible and open independent test
+confirmation.
+
 Prepare convolutional local-update primitives and ResNet-20/32/56 protocols early. Queue frozen
 runs opportunistically on authorized idle GPUs. Because BurstCCN already reports CIFAR-10 and
 ImageNet scaling, dataset scale alone is not novel. The oral-level target is a memorable joint