bibkey: shin2026iterated authors: Henry Shin year: 2026 title: “Iterated-sumset spectra: The complete exponent law and its rank geometry” doi: 10.48550/arXiv.2609.01690 claim: The question following Corollary 10.5 asks whether every four-point sumset cardinality is realizable at the minimum diameter of a maximal-cardinality model. strata_touched:
- D5/S0/Certificates/ShinFourPointCompression license: citation-only triage: anchor
Four-point sumset compression
Verified locator
DOI: https://doi.org/10.48550/arXiv.2609.01690
The pinned source is https://arxiv.org/html/2609.01690v1. Immediately before Corollary 10.5 it defines nu(h,4) as the least D for which all cardinalities of h-fold sumsets of four distinct integers occur in [0,D]. Corollary 10.5, equation (193), proves lower and upper bounds. The unnumbered paragraph after its proof, before Proposition 10.6, asks whether nu(h,4)=D_h^max for every h>=2. Here D_h^max=choose(h+2,2)+1 for odd h and choose(h+2,2) otherwise. The operation hA allows repetition and uses exactly h summands. This is an explicit question, without attributing an affirmative conjecture to Shin.
The source HTML read on 2026-09-13 has SHA-256 453e39bc6fc2773e50582eed117b46986f75341010fea1abc1db3a18bd506d60. The official abstract/version page https://arxiv.org/abs/2609.01690 listed only v1 in the same day’s check. A fresh public fetch of the pinned HTML returned HTTP 200. Only bibliographic information and the short question are paraphrased here; no source-paper text or code is redistributed.
Mathematical scope
The affirmative claim quantifies over every natural h>=2 and every finite four-element subset A of the integers. It asks for a four-element integer set B in the closed interval [0,D_h^max] with |hB|=|hA|. No additive-type, gcd, primitivity, reflection, or ordering requirement is imposed on A or B. The proposed negative instance is h=11, A={0,7,17,80}: |11A|=347, and that cardinality is absent from all four-element subsets of [0,79]. The Lean result consumes the full finite exclusion after translation normalization. It makes no assertion that nu(11,4)=80.
Bounded duplicate and literature search
At repository base e6ad855ec601b2195efc5abdf307ab282b31fd4d, searches of
D5, Library, Problems and Blueprint for 2609.01690, 2609.08915,
iterated.?sumset, and sumset found no matching compression result.
The sumset mention in GreedyThreeSumfreeTwoParameter concerns a different
three-sumfree construction. Pinned mathlib revision
db584cd6d46c92f209a44c0f1c829460d327499d has no matching compression theorem.
Its repeated-pointwise-addition, finite ordering, translation-cardinality,
binary-index, and bitwise lemmas are used directly in the proof.
The 2026-09-13 GitHub issue/PR searches for either paper ID within
trureturing returned zero. The global 2609.08915 issue search found one
arXiv feed issue. Formal Conjectures was checked at
a2f4a1bb12a28e04a969da78feefac7d1ce49565; the exact 2609.01690 code query
returned zero, and a fresh worker GitHub query returned total_count=0 with
incomplete_results=false. Public arXiv queries sumset spectra and
sumset compression, ordered newest first with size 50, included the Shin
paper as a positive control and found no newer matching solution.
These observations do not establish exhaustive publication priority.
Related source and limits
Zhang’s https://arxiv.org/html/2609.08915v1#S12.Thmtheorem5, Question 12.5, independently calls the same label-level question open and states that its weak-type theorem does not answer it. The source is catalogued as zhang2026sharp. Current source status, source-to-Lean fidelity, and absence of prior solutions are literature/review obligations, not consequences of a kernel axiom check. Independent review, canonical admission, freezing and delivery are separate from the local proof.