From e8b08e63ae29c7d97d8c05b9c55150be62161933 Mon Sep 17 00:00:00 2001 From: Dylan Roy Date: Tue, 13 Oct 2026 17:20:38 +0000 Subject: [PATCH 1/2] Math-#219 erratum 1 (XA-219-01..04): the correction-ratio reading is conditional Additive erratum answering Grok Bot agent 3's T8 read of Math-#219 (AMEND 5975882973, addendum 5975943358; routing 5976607568). - ERRATUM_20261013.md applies XA-219-01 verbatim as a reading of NOTE section 0, records XA-219-02..04 for the PR body, lists every nonauthor read, and records the two premises' status today. - README.md gains one pointer paragraph. - SOURCE_FILES.json gains the erratum's identity, README's new identity and an errata record. NOTE.md, RESULTS.json, SOURCE_MAP.json, every code file and the workflow are unchanged. Scientific effect NONE. Co-Authored-By: Claude Opus 5.5 Claude-Session: https://claude.ai/code/session_015wNj8LPTKXsaT68G3DgPPh --- .../ERRATUM_20261013.md | 120 ++++++++++++++++++ .../README.md | 4 + .../SOURCE_FILES.json | 21 ++- 3 files changed, 142 insertions(+), 3 deletions(-) create mode 100644 frontiers/cusp_coefficient_certified_20261001/ERRATUM_20261013.md diff --git a/frontiers/cusp_coefficient_certified_20261001/ERRATUM_20261013.md b/frontiers/cusp_coefficient_certified_20261001/ERRATUM_20261013.md new file mode 100644 index 000000000..19bc86b38 --- /dev/null +++ b/frontiers/cusp_coefficient_certified_20261001/ERRATUM_20261013.md @@ -0,0 +1,120 @@ +# Erratum (13 October 2026): the correction-ratio reading is conditional + +Object: CL-CU-CUSP-COEFFICIENT-20261001-v1, erratum 1. +Author: Anthropic Claude (the packet's author; Claude Code session `session_015wNj8LPTKXsaT68G3DgPPh`). +Scientific effect: NONE. No numerical, code or certificate change. + +`NOTE.md` stays unchanged as the merged record (v1.5, blob `b2866871`), and so does `RESULTS.json` (blob `8bbd0b0c`). +This file records the findings of the second-provider read of Math-#219: +- **The read.** Comment 5975882973 by Grok Bot agent 3, task T8 of main#229 5975667461, with its addendum 5975943358. +- **Its verdict.** AMEND: one additive wording erratum to `NOTE.md` (XA-219-01) and three PR-body items (XA-219-02..04). + It found no numerical or code defect. Its full replay on `main` `42f19d7` passed in `-B -S` and `-B -O -S`, and all six + mutants exit 1. + +## The finding + +`NOTE.md` §0 ("Consequences", second bullet) and the PR body call the manuscript's correction ratio "certified". What this +packet certifies is an enclosure of the ratio of the integral (CU.2) to `c_(3,24)`. Reading that ratio as the relative +correction `(c1/c) ell^(7/12)` of [CU] (CU.1) needs two premises that the packet consumes and does not review: Theorem CU, +and side24_v1's enclosure of `c_(3,24)`. `NOTE.md` §8 already states both limits; the §0 bullet is wider than they are. + +The PR body's `I^cand` bullet has the same shape. The identification of the certified intervals with the coefficient that +Math-#218 carries comes from #218 and [CU], not from this packet. + +## Erratum + +### XA-219-01: `NOTE.md` §0, "Consequences", second bullet + +This is the reviewer's text, applied verbatim. Read + +> **The manuscript's correction ratio is certified.** For the manuscript's field (`d = 3`, `L = 24`) the relative +> correction of [CU] (CU.1) is `(c1/c) ell^(7/12)`, with `c1/c_(3,24)` in `[-5.07106215, -5.07106213]` (Corollary S). + +as + +> **The manuscript's correction ratio is enclosed.** For the manuscript's field (`d = 3`, `L = 24`) the ratio of the +> integral (CU.2) to `c_(3,24)` lies in `[-5.07106215, -5.07106213]` (Corollary S). Its reading as the relative correction +> `(c1/c) ell^(7/12)` of [CU] (CU.1) is conditional on Theorem CU (candidate status, §0) and on side24_v1's enclosure of +> `c_(3,24)` (author-side, nonauthor review open, §8). + +The bullet's last sentence, "[CU] §0 states `c1/c ~ -5.071` for this field.", stays. + +### XA-219-02..04: the PR body of Math-#219 + +The PR body is not a repository file. Its readings are recorded here, with the reviewer's replacement text verbatim, and a +dated pointer at the top of the PR body leads here. + +- **XA-219-02 ("Consequences", second bullet).** The same overreach as XA-219-01. Read + + > **The manuscript's correction ratio is certified (v1.1).** For the manuscript's field (`d = 3`, `L = 24`), the + > relative correction of (CU.1) is `(c1/c) ell^(7/12)`, with `c1/c_(3,24)` in `[-5.07106215, -5.07106213]`. + + as + + > **The manuscript's correction ratio is enclosed (v1.1).** For the manuscript's field (`d = 3`, `L = 24`), the ratio + > of the integral (CU.2) to `c_(3,24)` lies in `[-5.07106215, -5.07106213]` (Corollary S). Its reading as the relative + > correction `(c1/c) ell^(7/12)` of (CU.1) is conditional on Theorem CU (Math-#207, candidate status) and on + > side24_v1's enclosure of `c_(3,24)` (author-side, nonauthor review open). + + The bullet's last sentence, "[CU] §0 quotes `c1/c ~ -5.071` for this field.", stays. + +- **XA-219-03 ("Consequences", fourth bullet).** Read + + > **`I^cand` is certified for Math-#218.** The certified `I^cand` intervals are the values of the `ell^(1/4)` + > coefficient of `nu_cand` that Math-#218 (Theorem T) carries: for the Gaussian kernel, and by Lemma S for the [P] + > field with `L >= 24`. + + as + + > **`I^cand` is enclosed for use by Math-#218.** The certified intervals enclose `I^cand = (3^(1/4)/2) c1` with `c1` + > the integral (CU.2): for the Gaussian kernel, and by Lemma S for the [P] field with `L >= 24`. Their reading as the + > `ell^(1/4)` coefficient of `nu_cand` in Math-#218 (Theorem T) is conditional on #218's identification and on [CU] + > (CU'.1) (candidate status). Neither is reviewed here. + + The bullet's last sentence, "#218 is consumed by nothing here.", is outside the reviewed quotation and stays. + +- **XA-219-04 ("Current disposition" and "Verification").** Both sections are historical, as of head `7a04873`. The merge + (`5a836a8`) and the packet on `main` supersede them; there `NOTE.md` is v1.5 (`d590cb8`, blob `b2866871`). In + particular: + - "[CU] is pinned at `826b8be` but is not on `main`" is outdated. [CU] (`frontiers/cusp_second_order_20261001/PROOF.md`, + blob `f6df5a73`) is on `main` since `566b1a1` and is pinned as `current_required` in `SOURCE_MAP.json` (`NOTE.md` §9). + - The PR body's list of nonauthor reads is incomplete. The full list follows. + +## Nonauthor reads of this packet + +- OpenAI / GPT-6 Astra Pro, R.1: ACCEPT (review 5378866830, on `357d0a1`). +- OpenAI / GPT-6 Astra Pro, R.2–R.6 with Lemma Q: ACCEPT_SCOPED (review 5384432735, on `7a04873`). +- OpenAI C54, three scoped reviews on `31542bb`: + - quadrature and tails: no blocking defect, with acceptance limited to the stated source and trajectory (review + 5384790625); + - constants, Lemma S and endpoint arithmetic: PASS_TECHNICAL (review 5384790842); + - Lemma A and its interval implementation: PASS_TECHNICAL (review 5384863435); + - the successor comparison with the merged head `29f26be`: 81 checks, zero failures (comment 5939714789). +- xAI / Grok, scope note on `3ac897a` (review 5379139737). Its NOT READY vote was a process hold, and 5975882973 records + it as superseded. +- xAI / Grok Bot agent 3, task T8: AMEND (5975882973, addendum 5975943358). This erratum answers it. + +## The two premises today + +Read on Math- `main` `b68d88b3` (13 October 2026). Both premises still have the status the erratum names: +- `frontiers/cusp_second_order_20261001/PROOF.md` (blob `f6df5a73`): "AUTHOR-SIDE PROOF CANDIDATE; NONAUTHOR REVIEW + REQUIRED". +- `coefficients/side24_v1/PROOF.md` (blob `44b66f04`): "author-side coefficient-evaluation candidate; nonauthor review + open". + +This file changes neither status. + +## What is unchanged + +- `NOTE.md`, `RESULTS.json`, `SOURCE_MAP.json` and every code file, with the same blobs; every number, interval, mutant + and pin. +- The records of the packets that consume this one. `frontiers/third_order_coefficient_20261001` pins `NOTE.md` at + `b2866871`. `frontiers/c2_torus_transfer_20261001` requires `RESULTS.json` at `8bbd0b0c` and records an earlier + `NOTE.md`, and `frontiers/cusp_torus_transfer_20261001` records the `NOTE.md` of `3ac897a`. + +`SOURCE_FILES.json` gains this file's identity, `README.md`'s new identity and an `errata` record. `README.md` gains one +pointer. + +The reader also asked that the same condition travel with "`c₁` certified for SIDE24" in V3. That text belongs to the V3 +manuscript, not to this packet. The routing note 5976607568 records that the project's V3 edit pack carries the condition +next to the SIDE24 `c₁` entry. diff --git a/frontiers/cusp_coefficient_certified_20261001/README.md b/frontiers/cusp_coefficient_certified_20261001/README.md index 8870c5bf7..7840312a4 100644 --- a/frontiers/cusp_coefficient_certified_20261001/README.md +++ b/frontiers/cusp_coefficient_certified_20261001/README.md @@ -1,5 +1,9 @@ # Certified cusp coefficient `c1` (Math-#207, (CU.2)), Gaussian kernel +**Erratum 1 (13 October 2026):** `ERRATUM_20261013.md` records that the manuscript's correction ratio is enclosed, and that +its reading as the relative correction of [CU] (CU.1) is conditional on Theorem CU and on side24_v1's `c_(3,24)`. It answers +the second-provider read 5975882973 (XA-219-01..04); no number or code changes. + `CL-CU-CUSP-COEFFICIENT-20261001-v1` (v1.1). Author-side certified numerics; scientific effect NONE. See `NOTE.md`. | `d` | `c1` | `c1 / c_(d,ref)` | diff --git a/frontiers/cusp_coefficient_certified_20261001/SOURCE_FILES.json b/frontiers/cusp_coefficient_certified_20261001/SOURCE_FILES.json index 8a329eb9e..f81d39944 100644 --- a/frontiers/cusp_coefficient_certified_20261001/SOURCE_FILES.json +++ b/frontiers/cusp_coefficient_certified_20261001/SOURCE_FILES.json @@ -1,6 +1,12 @@ { "object": "CL-CU-CUSP-COEFFICIENT-20261001-v1", "files": [ + { + "path": "ERRATUM_20261013.md", + "bytes": 7137, + "sha256": "85e17ef8879a974b0be3b2a321e9061c63df44139cf05584e5f56081893ced73", + "git_blob": "19bc86b3818adc02fa367cd5963d8ab5d0e71d68" + }, { "path": "NOTE.md", "bytes": 27264, @@ -9,9 +15,9 @@ }, { "path": "README.md", - "bytes": 774, - "sha256": "58c23e1062ee0d8d8d3f17af4dd6bfca76fc897ab51fc1dcc08481e4bf5b3c39", - "git_blob": "8870c5bf794baa0e9d1d2eca4f1ac2a509cf30e3" + "bytes": 1104, + "sha256": "14e427b7b9c0b2b823192e37cb2728036adae493dfdebef78d04b24232bf5bdf", + "git_blob": "7840312a485ce480149424430ec2508bb0852086" }, { "path": "RESULTS.json", @@ -85,5 +91,14 @@ "sha256": "85c330d82623dc89effed5a09b8d15a8986688f5eacc922f8b9b87754f9eef35", "git_blob": "31da88a031f3df682d1411415bbc1a9a7f591850" } + ], + "errata": [ + { + "file": "ERRATUM_20261013.md", + "date": "2026-10-13", + "found_by": "Grok Bot agent 3, Math-#219 comment 5975882973 (AMEND; no numerical or code defect) with addendum 5975943358, task T8 of main#229 5975667461", + "corrects": "NOTE.md section 0, Consequences, second bullet (XA-219-01): the correction ratio c1/c_(3,24) is enclosed, and its reading as the relative correction of [CU] (CU.1) is conditional on Theorem CU and on side24_v1's enclosure of c_(3,24); the PR-body readings XA-219-02..04 are recorded", + "note_unchanged": "NOTE.md bytes unchanged (blob b2866871c10cd0a0def41e10bdaf5ecbef18241c); RESULTS.json bytes unchanged (blob 8bbd0b0c69f6b29849fb2e3a3b2ad86974130987)" + } ] } From d517287c9f603d5dc59896261c3147ab15a06411 Mon Sep 17 00:00:00 2001 From: Dylan Roy Date: Tue, 13 Oct 2026 18:38:11 +0000 Subject: [PATCH 2/2] Math-#219 erratum 1: correct the side24_v1 premise status (Codex P2 4262546129) The "two premises today" section repeated side24_v1's out-of-date header ("nonauthor review open") as its current status. The same blobs (PROOF.md 44b66f04, ENCLOSURE.json 57af39a0) carry nonauthor reviews at the arithmetic-enclosure scope: xAI/Grok C1-C6 VERIFIED (main#65 5841269490) and reviews/side24_v1_coefficient_claude_20260929/REVIEW.md (ACCEPT at the arithmetic-enclosure scope); PROOF_INDEX.md records the coefficient-review work package as closed. The section now says so, keeps Theorem CU's candidate status, and notes that the reviewer's parenthetical stays verbatim. Only ERRATUM_20261013.md and its SOURCE_FILES.json identity change. NOTE.md, RESULTS.json, README.md, code and pins are unchanged. Co-Authored-By: Claude Opus 5.5 Claude-Session: https://claude.ai/code/session_015wNj8LPTKXsaT68G3DgPPh --- .../ERRATUM_20261013.md | 25 +++++++++++++------ .../SOURCE_FILES.json | 6 ++--- 2 files changed, 21 insertions(+), 10 deletions(-) diff --git a/frontiers/cusp_coefficient_certified_20261001/ERRATUM_20261013.md b/frontiers/cusp_coefficient_certified_20261001/ERRATUM_20261013.md index 19bc86b38..a9c3f95a6 100644 --- a/frontiers/cusp_coefficient_certified_20261001/ERRATUM_20261013.md +++ b/frontiers/cusp_coefficient_certified_20261001/ERRATUM_20261013.md @@ -96,13 +96,24 @@ dated pointer at the top of the PR body leads here. ## The two premises today -Read on Math- `main` `b68d88b3` (13 October 2026). Both premises still have the status the erratum names: -- `frontiers/cusp_second_order_20261001/PROOF.md` (blob `f6df5a73`): "AUTHOR-SIDE PROOF CANDIDATE; NONAUTHOR REVIEW - REQUIRED". -- `coefficients/side24_v1/PROOF.md` (blob `44b66f04`): "author-side coefficient-evaluation candidate; nonauthor review - open". - -This file changes neither status. +Read on Math- `main` `9844f785` (13 October 2026). These files are unchanged since `b68d88b3`. +- **Theorem CU.** `frontiers/cusp_second_order_20261001/PROOF.md` (blob `f6df5a73`) reads "AUTHOR-SIDE PROOF + CANDIDATE; NONAUTHOR REVIEW REQUIRED". No review of its proof is recorded under `reviews/` or in `PROOF_INDEX.md`; the + only `reviews/` record naming its packet, `cusp_workflow_protocol_20261006/SCOPE.md`, is an engineering note on its + checker workflow. So it keeps the candidate status the erratum names. +- **side24_v1's enclosure of `c_(3,24)`.** The header of `coefficients/side24_v1/PROOF.md` (blob `44b66f04`) still reads + "author-side coefficient-evaluation candidate; nonauthor review open", but that header is out of date. The same blobs, + `PROOF.md` `44b66f04` and `ENCLOSURE.json` `57af39a0`, have nonauthor reviews at the arithmetic-enclosure scope: + - xAI / Grok 4.7, with interfaces C1–C6 all VERIFIED (main#65 comment 5841269490); + - Anthropic Claude, a second provider: `reviews/side24_v1_coefficient_claude_20260929/REVIEW.md`, "ACCEPT at the + arithmetic-enclosure scope". + + `PROOF_INDEX.md` records that the reconciliation closes the coefficient-review work package. It also records that + reading the coefficient as the parent lifetime law keeps the parent's separate dependency and review boundaries. + +In the reviewer's replacement texts (XA-219-01 and XA-219-02), the parenthetical "nonauthor review open" follows that +header and `NOTE.md` §8. It stays verbatim as the reviewer's text and should be read with the reviewed status above. This +file changes neither status. (This section was corrected after Codex review finding 4262546129 on Math-#454.) ## What is unchanged diff --git a/frontiers/cusp_coefficient_certified_20261001/SOURCE_FILES.json b/frontiers/cusp_coefficient_certified_20261001/SOURCE_FILES.json index f81d39944..e9624bdeb 100644 --- a/frontiers/cusp_coefficient_certified_20261001/SOURCE_FILES.json +++ b/frontiers/cusp_coefficient_certified_20261001/SOURCE_FILES.json @@ -3,9 +3,9 @@ "files": [ { "path": "ERRATUM_20261013.md", - "bytes": 7137, - "sha256": "85e17ef8879a974b0be3b2a321e9061c63df44139cf05584e5f56081893ced73", - "git_blob": "19bc86b3818adc02fa367cd5963d8ab5d0e71d68" + "bytes": 8430, + "sha256": "c6e4d0ae3854b27f31b7165ccc376d070c85093a7e442d8b951744013b9b0be1", + "git_blob": "a9c3f95a6e31cd1924f4952d6092d848bd80f8ef" }, { "path": "NOTE.md",