Skip to content

Exclude compensated P104 fusion readouts (0% κ change) - #7338

Open
ToikY wants to merge 3 commits into
CrocSwap:mainfrom
ToikY:codex/p104-compensated-fusion-readout-bound
Open

ToikY wants to merge 3 commits into
CrocSwap:mainfrom
ToikY:codex/p104-compensated-fusion-readout-bound

Conversation

@ToikY

@ToikY ToikY commented Nov 11, 2026

Copy link
Copy Markdown
Contributor

Change and mathematical scope

NO VERIFIED GAIN. Exact conditional κ change 0%. On exact #7334, the one-helper fusion screen exhausted 3,422 paid pairs. It fixes the original paid itineraries except the fused helper pair, and uses one maximal public scalar component with at most 24 events and 20 roles, one low-chart peer, and a shared legal retiming cut. Of 456 geometry survivors, all 1,197 strict paired keys fail exterior support. Allowing the virtual helper’s owned output to be cleared and restoring the dirty public helper with a nonzero scalar leaves 820 nonlocal forward failures and 377 one-peer readout failures. Independent integer rank certificates exclude all 43 forward-feasible extensions to the full high-chart readout bank. Zero cases remain unresolved within scope.

The complete 17-copy signed invoice has 110,519 paid and 22,461 productive positions, surplus 1,424 and no empty credit. The one-edit optimistic upper is 0.380712103179 raw units, rounded upward, and an extra paid rank-one position kills it. Actual compatible saving is zero. The peer benchmark is κ = 0.00364665195644 on #7334 at aaeec8c, independently reproduced from all four registered stages. Its separate gain over #7314 is exactly 0.000000430519994.

Replay

python3.12 -I -B research/p104-compensated-fusion-readout-bound/replay.py
python3.12 -B scripts/research_package.py --package research/p104-compensated-fusion-readout-bound
python3.12 -B scripts/check_research_commit.py --package research/p104-compensated-fusion-readout-bound --check-modes --check-source-history

All 17 pinned receipts reproduce, with known-good #319 reconstruction and negative controls. Registered replay, committed bytes/modes/source history, Git-free export and six navigation tests passed locally. No literal candidate passed support. Larger components, different clocks or paid itineraries, and multiple low peers remain outside scope. The branch retains the #7257 replay prerequisites.

Credit and overlap

#7334 supplies the closed-block fusion and restored dirty return, #7314 its parent, #7113 the fixed signed caller, and #7298 the directed tariff. #7308 supplies retained owned-MOVE evidence and related separating-coordinate methods. #6529 and #6765 are compensated-fusion and resident-response precedents on earlier words, and #7312/#7329 have different fixed-contact scopes. OpenAI Codex and subagents materially assisted.

Clear1nsight and others added 3 commits November 11, 2026 15:39
This is the exact documentation prerequisite retained by CrocSwap#7156. It points the onboarding check to the selected CrocSwap#7000/CrocSwap#7001 review. Credit to Rohan Arun and the source authors.
The PR CrocSwap#6292 rejection census and package plan are statements about 2485805. Load its exact sources and planner in a private sparse checkout, keeping all source hashes, 19 expected rejections, nine package selections and the empty-plan control. Subsequent edits to the live Makefile and CI guide no longer invalidate this historical check.

The original audit is by unpaidmoon, building on Dennis’s historical replay gate. The pinned repository view follows the precedent of Matarener’s PR CrocSwap#5434. Repair prepared with substantial OpenAI Codex assistance.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants