Repository navigation
Independently audit native equality-cut signs and caller contracts - #5690
Merged
Mister-Meeseeks merged 1 commit intoOct 26, 2026
Merged
Conversation
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
PR #5588 requested independent review of its signed stabilizer tariff and full-word/cached-history/endpoint contracts. This package checks
PROOF.md§§3–5 at97db3fb45cd93a350e300781d75440d14e6a1f73: the claims hold under the stated cut-preserving hypotheses; no defect was found in that scope. A separate trace derivation makes the Pauli-sign cancellation explicit and proves the bound for arbitrary n, width, chronological feedback and complete initial availability J.The independently written exact checks cover 366 signed ordered Pauli image frames (not all Clifford unitaries), 54 literal projectors, Bell-sign and inter-column endpoint controls, one 18-operation dirty-feedback word and all 19 prefixes, ten paid-update cases, joint cache rank and nonunitary code controls. In particular, duplicate cached responses have joint rank1 rather than2, and preserving cut rank alone does not preserve the unit error floor.
Validation: registered Python3.12 replay and isolated optimized replay; committed-byte/source-history checks; deterministic 13-file export and integrity check; fresh Git-free optimized replay with identical receipt; altered-source and altered-receipt rejection controls. See
research/native-equality-independent-audit/VERIFICATION.md. Current main was refreshed before submission and remains the pinned basedf0137efd35e11b4445734bf612bdd4c7ebc330c.Scope excludes the source profile algorithm (§6), row-packing discussion (§7), actual P104 availability extraction, supplier/physical replay, and the full multiplication theorem. No κ improvement or frontier change; selected P104 κ=62009/20000000 retains its five premises. This is an agent-assisted independent derivation and implementation, not independent human or proof-kernel acceptance.
Canonical claim/handoff: #5582 (comment)
Original review request: #1166 (comment)
Ali Romman retains credit for the source theorem and its cached-boundary/endpoint limitations; classical rank/stabilizer theory and all source predecessor attribution are retained. New audit prepared for Hiroki KUMAZAKI with substantial OpenAI Codex assistance, including delegated proof, contract-check and sign review. Apache-2.0.