diff --git a/docs/RESEARCH_INDEX.md b/docs/RESEARCH_INDEX.md index b3a42ed88..eb8b1f307 100644 --- a/docs/RESEARCH_INDEX.md +++ b/docs/RESEARCH_INDEX.md @@ -18,7 +18,7 @@ Checked on 12 October 2026 against main `8977619e86d2f395b8e4947cf9a5e34082cc2c2 5. **P15 combinatorics: restricted theorem with a scoped nonauthor review.** [Theorem F](https://github.com/d6g8k5htny-coder/Math-/blob/79cb314d5b9f6b8b410bfd0df200414314f6b8c7/frontiers/full_price_20260924/PROOF.md) (author OpenAI) holds on its realized disjoint-block family with demands `d_i>=2`, for every independent probability vector, under its price and palette hypotheses; an xAI Grok 4.7 [nonauthor review](https://github.com/d6g8k5htny-coder/Math-/blob/79cb314d5b9f6b8b410bfd0df200414314f6b8c7/reviews/p15_full_price_nonauthor_20260926/REVIEW.md) accepts it at exactly that scope. Arbitrary downsets and an unrestricted P15 result are outside it; see [P15 combinatorics](#p15-combinatorics). -6. **Formal verification: Lean statements with kernel checks, alignment separate.** The [cap-first-exit packet](https://github.com/d6g8k5htny-coder/Math-/blob/79cb314d5b9f6b8b410bfd0df200414314f6b8c7/frontiers/cap_first_exit_lean_20261002/README.md) (author Anthropic Claude) has 152 Lean targets in its [manifest](https://github.com/d6g8k5htny-coder/Math-/blob/79cb314d5b9f6b8b410bfd0df200414314f6b8c7/frontiers/cap_first_exit_lean_20261002/MANIFEST.json). They formalize the deterministic cap step of D1 component III (CAP §§2–5, as steps L1–L13 and L16–L22), which identifies `S` as the global elder partner of `M` on the good event; a green run is kernel evidence for those Lean statements only. Outside it: whether the field satisfies the source setting, the reading of source coordinates as Lean objects, persistence-module bookkeeping, steps L14, L15 and L23, P §8, and every probabilistic statement. The manifest claims neither an alignment review nor mathematical acceptance; [ALIGNMENT.md](https://github.com/d6g8k5htny-coder/Math-/blob/79cb314d5b9f6b8b410bfd0df200414314f6b8c7/frontiers/cap_first_exit_lean_20261002/ALIGNMENT.md) records the earlier scoped read of the v1.4 statements and what a non-Claude review must cover. Main's own formal layer is under [Open work](#open-work). +6. **Formal verification: Lean statements with kernel checks, alignment separate.** The [cap-first-exit packet](https://github.com/d6g8k5htny-coder/Math-/blob/79cb314d5b9f6b8b410bfd0df200414314f6b8c7/frontiers/cap_first_exit_lean_20261002/README.md) (author Anthropic Claude) has 152 Lean targets in its [manifest](https://github.com/d6g8k5htny-coder/Math-/blob/79cb314d5b9f6b8b410bfd0df200414314f6b8c7/frontiers/cap_first_exit_lean_20261002/MANIFEST.json). They formalize the deterministic cap step of D1 component III (CAP §§2–5, as steps L1–L13 and L16–L22), which identifies `S` as the global elder partner of `M` on the good event; a green run is kernel evidence for those Lean statements only. Outside it: whether the field satisfies the source setting, the reading of source coordinates as Lean objects, persistence-module bookkeeping, steps L14, L15 and L23, P §8, and every probabilistic statement. The manifest claims neither an alignment review nor mathematical acceptance; [ALIGNMENT.md](https://github.com/d6g8k5htny-coder/Math-/blob/79cb314d5b9f6b8b410bfd0df200414314f6b8c7/frontiers/cap_first_exit_lean_20261002/ALIGNMENT.md) records the earlier scoped read of the v1.4 statements and what a non-Claude review must cover. Main's own formal layer is under [Open work](#open-work). Read after that check, at Math- `a0323b94`: the statement-only [Lean specifications](https://github.com/d6g8k5htny-coder/Math-/blob/a0323b94ae54a78f3d2878dc9c358dc40cdfe1dc/specifications/lean_statements_20261010/README.md) of registered claims and their [proof packet](https://github.com/d6g8k5htny-coder/Math-/blob/a0323b94ae54a78f3d2878dc9c358dc40cdfe1dc/specifications/lean_statement_proofs_20261011/README.md) (author Anthropic Claude) landed through Math- [#419](https://github.com/d6g8k5htny-coder/Math-/pull/419), [#422](https://github.com/d6g8k5htny-coder/Math-/pull/422), [#423](https://github.com/d6g8k5htny-coder/Math-/pull/423), [#434](https://github.com/d6g8k5htny-coder/Math-/pull/434) and [#439](https://github.com/d6g8k5htny-coder/Math-/pull/439) on 12 October, after the Math- commit checked above; the packet holds 75 theorems, each kernel-checked by a hosted pull-request run, of 74 of the register's 141 Props, including `FullTransformedPriceBudget`, the Prop the register gives for Theorem F's display (F2) (see [P15 combinatorics](#p15-combinatorics)), with alignment PENDING_INDEPENDENT_REVIEW, organizational-independence credit 0 and scientific effect NONE; each theorem is evidence about its Lean statement only and establishes neither Theorem F nor any other parent claim, and the register's labels are unchanged (every Prop and structure of the statement packet stays `specified` and no claim-level label changes; the register adds the packet's theorems as entries of kind `theorem` labelled `proved`). 7. **Zeta-zero heat-flow folds and a CUE short-bar control: exploratory packet; scientific effect NONE.** The [zeta fold exploration packet](../experiments/zeta_fold_exploration/README.md) (author Anthropic Claude), landed through [main #373](https://github.com/d6g8k5htny-coder/main/pull/373) after the main commit checked above, checks an external note's proposed links between zeta zeros and the project's short-bar law: near the collision time, nondegenerate zero collisions of the de Bruijn–Newman heat flow follow the generic cubic fold, and a finite-window short-bar statistic of a Hardy-Z primitive is compared with CUE and one-dimensional Gaussian controls. It makes no claim of progress toward the Riemann Hypothesis. OpenAI/Codex nonauthor reads of its slices A/D (mathematics, wording) and B (citations) returned AMEND, and their delta readbacks returned scoped PASS; independent numerical reproduction (slice C) has not been performed. @@ -114,7 +114,7 @@ The shared-coordinate packets drop the disjoint-block and macro-clutter structur A [complete scoped review](https://github.com/d6g8k5htny-coder/Math-/blob/79cb314d5b9f6b8b410bfd0df200414314f6b8c7/reviews/p15_overlap_tail_claude_20260928/REVIEW.md) by an Anthropic Claude session records ACCEPT on every listed interface of both packets and found no defect. The reviewer's provider differs from the OpenAI author's, but all lanes share one project account, so organizational-independence credit is 0. An earlier [OpenAI same-provider read](https://github.com/d6g8k5htny-coder/Math-/pull/118#pullrequestreview-5344833684) of the bipartite packet found no blocker. The review's remark R1, extending the sub-1/2 factor to bipartite instances whose active blocks all have capacity at least 2, is the reviewer's result; it was not inserted into either source proof. Sources (Math- #118, #122) and review (Math- #123) landed together through [Math- PR #124](https://github.com/d6g8k5htny-coder/Math-/pull/124) with its [integration record](https://github.com/d6g8k5htny-coder/Math-/blob/79cb314d5b9f6b8b410bfd0df200414314f6b8c7/docs/integration/2026-09-28-reviewed-mathematics.md). Neither packet supplies arbitrary downsets or an unrestricted prize result. -Still open: arbitrary downsets and the unrestricted prize; overlap patterns outside these two classes; optimal cover factors for the overlap classes; and a source-bound nonauthor theorem review of the 16/27 theorem, which the claim manifest holds as `HOLD_WITH_DOMAIN`. At Math- `79cb314d` the downstream graph still lists `math.p15-full-price` as `AUTHOR_SIDE_CANDIDATE` ("review open"); the Math- #167 record proposes aligning it with main `STATUS.md` row D6, and that step has not been executed. Entries are listed after their publication is verified; this section was checked at Math- `79cb314d`. Lean work on Theorem F in open draft Math- pull requests is not listed here; its formal target is under [Open work](#open-work). +Still open: arbitrary downsets and the unrestricted prize; overlap patterns outside these two classes; optimal cover factors for the overlap classes; and a source-bound nonauthor theorem review of the 16/27 theorem, which the claim manifest holds as `HOLD_WITH_DOMAIN`. At Math- `79cb314d` the downstream graph still lists `math.p15-full-price` as `AUTHOR_SIDE_CANDIDATE` ("review open"); the Math- #167 record proposes aligning it with main `STATUS.md` row D6, and that step has not been executed. Entries are listed after their publication is verified; this section was checked at Math- `79cb314d`. Read after that check, at Math- `a0323b94`: Lean work on Theorem F landed after the Math- commit checked above, through [Math- #419](https://github.com/d6g8k5htny-coder/Math-/pull/419) (the [statement register](https://github.com/d6g8k5htny-coder/Math-/blob/a0323b94ae54a78f3d2878dc9c358dc40cdfe1dc/specifications/lean_statements_20261010/REGISTER.json)) and [Math- #422](https://github.com/d6g8k5htny-coder/Math-/pull/422) and [#423](https://github.com/d6g8k5htny-coder/Math-/pull/423) (the [proof packet](https://github.com/d6g8k5htny-coder/Math-/blob/a0323b94ae54a78f3d2878dc9c358dc40cdfe1dc/specifications/lean_statement_proofs_20261011/README.md)), all merged 12 October with packet author Anthropic Claude, and when checked on 12 October no open Math- pull request carried Lean work on it; the packet has theorems, each kernel-checked by a hosted pull-request run, of 19 of the 20 Props the register lists for claim `p15-full-price`, all but the predicate `IsProperDownset`, among them `FullTransformedPriceBudget` (the Prop the register gives for Theorem F's display (F2)), `RhoStarEnclosure` (F3), `CoordinatewiseConcavity` (the concavity the source derives from (F4)–(F5), which themselves are not specified) and the Props of (F6)–(F8), (F12) and (F13); alignment is PENDING_INDEPENDENT_REVIEW, organizational-independence credit 0 and scientific effect NONE, each theorem is evidence about its Lean statement only and does not establish the informal Theorem F, the register's labels are unchanged (the claim and each of its 20 Props stay `specified`; the register adds the packet's 19 theorems to that row as entries labelled `proved`, beside 7 finite-arithmetic theorems of the open main draft [main #358](https://github.com/d6g8k5htny-coder/main/pull/358) that it also labels `proved` and records as not on main and not Theorem F), and the theorems of `RhoStarEnclosure` and `RhoStarBelowSixSevenths` change no status recorded for the formal target for Theorem F's constant `ρ*`; the register keeps both Props `specified`. Separately, the [Open work](#open-work) table still lists that target. ## Open work