docs: sync h2 output-cloud consumer boundary#314
Merged
DeliciousBuding merged 1 commit intoMay 24, 2026
Conversation
DeliciousBuding
deleted the
docs/h2-output-cloud-consumer-boundary-sync-20260525
branch
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request updates the documentation for the 'H2 output-cloud geometry' research candidate, providing detailed performance metrics and clarifying its status as a research-side component rather than a platform feature. The review feedback identifies a duplication of this entry in the reproduction status table with inconsistent status labels and suggests consolidating these entries for a single source of truth. Additionally, there is a recommendation to align the list of technical exclusions in the product bridge documentation with the evidence documentation to ensure consistency.
| | True known-split mechanisms | `hold-weak` | MNIST/DDPM raw-loss and x0 scouts are weak. Tiny overfit final-layer gradient norm was positive only on the extreme `8 / 64` target, weakened at `16 / 64`, and a more optimistic `64 / 64` oracle gradient-prototype alignment follow-up is effectively random (`AUC = 0.500977`, `ASR = 0.562500`, zero low-FPR recovery). Fashion-MNIST DDPM now has three weak clean-split scouts: fixed-timestep PIA-style loss (`AUC = 0.535889`, `TPR@1%FPR = 0.03125`), SimA single-query score-norm (`AUC = 0.515137`, zero low-FPR recovery), and score-Jacobian sensitivity (`AUC = 0.511719`, zero low-FPR recovery). The Beans member-LoRA denoising-loss scout repaired pseudo-membership semantics by creating an exact `SD1.5 + Beans-member LoRA` target, but the internal conditional denoising-loss score is weak (`AUC = 0.414400`, reverse `0.585600`, `TPR@1%FPR = 0.080000`) and parameter-delta sensitivity is also weak (`AUC = 0.512000`). Do not run more final-layer gradient norm/cosine variants, Fashion-MNIST timestep/seed/`p`-norm/perturbation/norm/packet-size sweeps, or Beans LoRA train-step/rank/resolution/prompt/timestep/layer matrices by default. See [fashion-mnist-ddpm-score-jacobian-sensitivity-20260514.md](fashion-mnist-ddpm-score-jacobian-sensitivity-20260514.md), [fashion-mnist-ddpm-sima-score-norm-20260514.md](fashion-mnist-ddpm-sima-score-norm-20260514.md), [beans-lora-delta-sensitivity-20260513.md](beans-lora-delta-sensitivity-20260513.md), [beans-lora-member-denoising-loss-scout-20260513.md](beans-lora-member-denoising-loss-scout-20260513.md), [fashion-mnist-ddpm-pia-loss-scout-20260513.md](fashion-mnist-ddpm-pia-loss-scout-20260513.md), [tiny-known-split-gradient-prototype-alignment-20260513.md](tiny-known-split-gradient-prototype-alignment-20260513.md), [gradient-norm-stability-gate-20260512.md](gradient-norm-stability-gate-20260512.md), and [tiny-overfit-gradient-norm-scout-20260512.md](tiny-overfit-gradient-norm-scout-20260512.md). | | ||
| | Black-box `H2 response-strength` | candidate-only | Positive-but-bounded DDPM/CIFAR10 candidate: frozen cutoff-0.50 lowpass follow-up passed, and raw H2 recovered strict-tail signal on the fresh packet. SD/CelebA text-to-image transfer is blocked by protocol mismatch. The frozen SD/CelebA image-to-image micro-packet is runnable, but H2 logistic does not beat the same-cache simple distance comparator, so H2 is not promoted beyond candidate-only. A separate simple-distance line now has bounded single-asset evidence: first 10/10 packet `AUC = 0.92`, non-overlapping 10/10 packet `AUC = 0.99` with 9/10 TP at 0 FP, and non-overlapping 25/25 admission packet `AUC = 0.8768`, `ASR = 0.84`, 11/25 TP at 0 FP. This is not a conditional-diffusion generalization or a `recon` product replacement. See [black-box-response-strength-preflight.md](black-box-response-strength-preflight.md), [h2-lowpass-followup-contract.md](h2-lowpass-followup-contract.md), [h2-cross-asset-contract-preflight.md](h2-cross-asset-contract-preflight.md), [h2-image-to-image-contract.md](h2-image-to-image-contract.md), [h2-img2img-micro-result.md](h2-img2img-micro-result.md), [h2-img2img-simple-distance-review.md](h2-img2img-simple-distance-review.md), [h2-img2img-simple-distance-stability-result.md](h2-img2img-simple-distance-stability-result.md), and [h2-img2img-simple-distance-admission-result.md](h2-img2img-simple-distance-admission-result.md). | | ||
| | Black-box `H2 output-cloud geometry` | candidate-only | CPU-only review on the existing H2 response cache found a strong output-output geometry signal that excludes seed-to-output distance (`AUC = 0.961529`, `ASR = 0.900391`, `TPR@1%FPR = 0.333984`, `TPR@0.1%FPR = 0.117188`). Seed `177` remains stable (`AUC = 0.961048`), and label-shuffle sanity returns random-level (`AUC = 0.507595`). This is not admitted because the source cache used class-ordered sample offsets, so a reseeded or interleaved order-control cache is required before promotion. Do not expand into KDE, shadow-density, repeat-count, same-cache feature sweeps, Platform schema, Runtime runner, or bundle rows. See [h2-output-cloud-geometry-20260525.md](h2-output-cloud-geometry-20260525.md). | | ||
| | Black-box `H2 output-cloud geometry` | candidate-only | Output-output geometry is now a strong H2 Research-side candidate after seed-offset control, not a Platform/Runtime row. The existing `512 / 512` cache review gives `AUC = 0.961529`, `ASR = 0.900391`, `TPR@1%FPR = 0.333984`, `TPR@0.1%FPR = 0.117188`; seed `177` remains stable (`AUC = 0.961048`), and label-shuffle sanity returns random-level (`AUC = 0.507595`). The bounded `256 / 256` shared-position order-control scout preserves the signal (`AUC = 0.967819`, `TPR@1%FPR = 0.410156`, `TPR@0.1%FPR = 0.132812`) with random-level label shuffle (`AUC = 0.464066`), and the same controlled boundary at seed `177` remains strong (`AUC = 0.956192`, `TPR@1%FPR = 0.285156`, `TPR@0.1%FPR = 0.109375`) with random-level label shuffle (`AUC = 0.484070`). Class-ordered seed offset is no longer a sufficient explanation, but this remains a single H2 response-cache geometry candidate, not a second public asset or consumer contract. Do not expand into KDE, shadow-density, repeat-count, same-cache feature sweeps, full `512 / 512` reruns, Platform schema, Runtime runner, or bundle rows. See [h2-output-cloud-geometry-20260525.md](h2-output-cloud-geometry-20260525.md). | |
There was a problem hiding this comment.
The track Black-box H2 output-cloud geometry is duplicated in this table (see line 35). Additionally, the status label used here (candidate-only) is inconsistent with line 35 (hold-candidate), and neither appears in the Status Stages definitions (lines 8-26). Please consolidate these entries to ensure a single source of truth and use a status label that aligns with the defined stages (e.g., deferred-candidate).
| 当前产品桥接决策记录在 | ||
| [h2-simple-distance-product-bridge-comparison.md](h2-simple-distance-product-bridge-comparison.md)。 | ||
|
|
||
| 当前 H2 output-cloud geometry 状态：这是 Research 侧强机制候选，不是 Platform/Runtime 行。既有 H2 `512 / 512` response-cache review、`256 / 256` shared-position order-control scout 和 seed `177` stability scout 都保持强信号，且 label-shuffle 回到随机级；但该结果仍只是 H2 response-cache geometry 候选，不是第二公开资产、消费合约或产品准入证据。不要新增 Platform schema、Runtime runner、UI 类型、bundle row、同 cache feature sweep 或完整 `512 / 512` 补跑。参见 |
There was a problem hiding this comment.
For consistency with the detailed evidence documentation in docs/evidence/reproduction-status.md, consider including the specific technical exclusions (KDE, shadow-density, repeat-count) in the list of restricted expansions.
Suggested change
| 当前 H2 output-cloud geometry 状态：这是 Research 侧强机制候选，不是 Platform/Runtime 行。既有 H2 `512 / 512` response-cache review、`256 / 256` shared-position order-control scout 和 seed `177` stability scout 都保持强信号，且 label-shuffle 回到随机级；但该结果仍只是 H2 response-cache geometry 候选，不是第二公开资产、消费合约或产品准入证据。不要新增 Platform schema、Runtime runner、UI 类型、bundle row、同 cache feature sweep 或完整 `512 / 512` 补跑。参见 | |
| 当前 H2 output-cloud geometry 状态：这是 Research 侧强机制候选，不是 Platform/Runtime 行。既有 H2 `512 / 512` response-cache review、`256 / 256` shared-position order-control scout 和 seed `177` stability scout 都保持强信号，且 label-shuffle 回到随机级；但该结果仍只是 H2 response-cache geometry 候选，不是第二公开资产、消费合约或产品准入证据。不要扩展到 KDE、shadow-density、repeat-count、同 cache feature sweep、完整 `512 / 512` 补跑，或新增 Platform schema、Runtime runner、UI 类型和 bundle row。参见 |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Checks