REVIEW 1 major objections 4 minor 2 cited by
Additive sumset sizes with tetrahedral differences
T0 review · 1 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash
Pith's one-line read For every positive integer $h$ and every $i_0\in[0,h-1]$, the four-element set $\{0,1,h+1,(h+1)(h+1-i_0)\}$ has $h$-fold sumset size $\binom{h+3}{3}-\binom{i_0+2}{3}$, proving that all tetrahedral-difference 'popular' sumset sizes exist.
desk verdict Proves the existence of the popular tetrahedral-difference sumset sizes for k=4 with explicit h-adic sets; the main counting argument is sound, though one edge case has an unproved (but true) B_h assertion. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The engine is a layer decomposition of the sumset. Put $B=\{0,1,h+1\}$ and $c=(h+1)(h+1-i_0)$, so $A=B\cup\{c\}$. Then $hA=\bigcup_{i=0}^h L_i$ with $L_i=(h-i)B+ic$. Each $L_i$ is itself a union of integer intervals of length at most $h$, and the parameter $c$ is chosen so that $L_i$ and $L_{i+1}$ overlap exactly for $i<i_0$, with overlap size $\binom{i_0+1-i}{2}$. Summing these overlaps gives the correction $\sum_{i=0}^{i_0}\binom{i_0+1-i}{2}=\binom{i_0+2}{3}$, leaving the tetrahedral difference $\binom{h+3}{3}-\binom{i_0+2}{3}$. The endpoint $i_0=0$ uses unique representation: $A=\{0,1,h+1,(h+1)^2\}$ should be a $B_h$-set, meaning every element of $hA$ has only one unordered representation as a sum of $h$ elements.
What would settle it
Count the distinct $h$-fold sums of $\{0,1,h+1,(h+1)^2\}$ for $h=4$: the theorem says all 35 unordered multisets give distinct values, so finding two equal sums would refute the endpoint proof. Independently, count $|hA|$ for $h=4$, $i_0=2$, where the formula predicts $\binom{7}{3}-\binom{4}{3}=31$; a direct enumeration yielding any other number would show the overlap-counting argument fails.
Extended reading notes
Core claim
The central claim is a constructive existence theorem. For every $h\ge 1$, the $h$ integers $\binom{h+3}{3}-\binom{i_0+2}{3}$ for $i_0=0,\dots,h-1$ all lie in $\mathrm{RZ}(h,4)$, the set of sizes of $h$-fold sumsets of four-element sets of integers. The witness is the explicit $h$-adically defined set $A=\{0,1,h+1,(h+1)(h+1-i_0)\}$. The proof decomposes $hA$ into the layers $(h-i)\{0,1,h+1\}+ic$, counts the sizes of the layers, and subtracts only the adjacent overlaps, whose total turns out to be $\binom{i_0+2}{3}$; the remaining layers are disjoint. Thus the tetrahedral-difference sizes are not merely compatible with the general bounds on $|hA|$; they are attained.
Load-bearing premise
The load-bearing premise is a missing unique-representation check: for $i_0=0$ the proof needs $\{0,1,h+1,(h+1)^2\}$ to have exactly one unordered representation of every $h$-fold sum, but it states this as a 'B4' property and gives no proof.
Editorial extensions
If this is right
- For each $h$, the $h$ tetrahedral-difference sizes $\binom{h+3}{3}-\binom{i_0+2}{3}$, $i_0=0,\dots,h-1$, are all attainable by four-element sets, so $\mathrm{RZ}(h,4)$ contains at least $h$ distinct values.
- The maximal $h$-fold sumset size $\binom{h+3}{3}$ is attained by $A=\{0,1,h+1,(h+1)^2\}$, so the upper bound in the general inequality for $|hA|$ is sharp for four-element sets.
- Together with the exact triangular-difference result for $k=3$, this gives a growing family of computed 'popular' sumset sizes and provides a concrete starting point for determining the full sets $\mathrm{RZ}(h,k)$.
- The one-parameter family of constructed sets can be used directly in computations of $hA$ for large $h$, since the counting problem reduces to a calculation indexed by $i_0$.
Reading between the lines
- A natural extension, not attempted in the paper, is to use the same layer-and-overlap scheme with $B$ of size $k-1$ plus one large element; this could plausibly produce sumset sizes that are differences of higher-dimensional figurate numbers, such as pentatope numbers for $k=5$.
- The theorem establishes existence but not the experimentally suggested concentration. One could sample four-element sets and compare the empirical frequency of these tetrahedral-difference sizes with other sizes in the allowed range to test whether the predicted distribution actually holds.
- The explicit construction corresponds to the special parameter $p=1+(i_0-1)(h+1)$ in the larger family $A=\{0,1,h+1,h^2+h+1-p\}$ raised as an open problem, so it offers a natural testbed for that problem.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper studies the set RZ(h,k) of possible sizes of h-fold sumsets of k-element integer sets. For k=4, numerical experiments suggested that the 'popular' sumset sizes are the differences of tetrahedral numbers C(h+3,3)-C(i0+2,3) for i0 in [0,h-1]. The main theorem constructs, for every h≥1 and every i0 in [0,h-1], an explicit four-element set A={0,1,h+1,(h+1-i0)(h+1)} and proves that |hA| = C(h+3,3)-C(i0+2,3). The proof decomposes hA into sets L_i indexed by the number of occurrences of the largest element, analyzes these as unions of intervals in residue classes modulo h+1, and computes the sizes via intersection counts. The i0=0 case is treated separately by asserting a B4-set property.
Significance. If the proof is completed, the paper settles the existence part of the tetrahedral-difference conjecture for k=4 and provides an explicit h-dependent construction for each popular sumset size. This is a modest but genuine contribution to a recently active problem, complementing the known triangular-difference result for k=3. The main counting argument for i0≥1 is self-contained and elementary, and the construction is explicit. The paper does not determine the full set RZ(h,4), only a distinguished subset, so its significance is proportionate to that scope.
major comments (1)
- [Section 2, h≥2 and i0=0 paragraph] The proof of the i0=0 case is incomplete as written. The set A={0,1,h+1,(h+1)^2} is called 'a B4-set and so |hA| = C(h+3,3)'. This inference is invalid: the B4 property controls unique representations in 4A, whereas the displayed equality is exactly the assertion that A is a B_h-set, i.e. that every integer in hA has a unique representation up to permutation. The B_h property is in fact true for this set — reducing any collision b+c(h+1)+d(h+1)^2 = b'+c'(h+1)+d'(h+1)^2 modulo h+1 gives b=b', then c=c', then d=d' — but this argument is not supplied and the statement is mislabeled. Since i0=0 belongs to the theorem's range, this is a load-bearing gap that must be repaired by adding the B_h proof.
minor comments (4)
- [Section 2, h=1 paragraph] The theorem requires i0∈[0,h-1], so for h=1 the only case is i0=0, but the proof writes 'i0=1'. The set A={0,1,2,4} that follows is correct for i0=0, so this is a typographical error that should be corrected.
- [Section 2, disjointness threshold display] The displayed equivalence 'max(L_i)<min(L_{i+1}) if and only if i≥1+floor((h-c)/(h+1))=i0' is arithmetically incorrect: (h-c)/(h+1)=i0-1-1/(h+1), so 1+floor((h-c)/(h+1))=i0-1, not i0. The preceding inequality correctly gives i>i0-1, hence i≥i0, and the subsequent argument uses the threshold i0, so the conclusion is unaffected; the display should nevertheless be corrected or simplified.
- [References] References [7] and [8] both list the arXiv identifier 2411.02365; since they are described as different papers (one in Fibonacci Quarterly and one in Acta Mathematica Hungarica), the identifiers should be checked and corrected.
- [Throughout] The phrase 'B4-set' in the i0=0 paragraph should read 'B_h-set'; the same correction is part of the major comment above and should be applied consistently.
Circularity Check
No significant circularity: the theorem is proved by a direct, self-contained construction.
full rationale
The paper's central claim is an existence theorem: for every h and i0 in [0,h-1], the explicitly defined set A = {0, 1, h+1, (h+1-i0)(h+1)} has |hA| equal to the stated tetrahedral difference. The proof in Section 2 computes |hA| directly by decomposing hA into layers L_i, counting the sizes of the layers via the elementary fact that B = {0,1,h+1} is a B_n-set for n ≤ h, and then computing pairwise intersections of consecutive layers to obtain the exact correction term. All of these steps are carried out in the paper with explicit formulas; no fitted parameter is later relabeled as a prediction, and no load-bearing result is imported from the author's earlier work. The citations to Nathanson [8], Nathanson [11], and O'Bryant [13] are used only for context, motivation, and experimental background, not as premises of the proof. The only weakness in the written proof is the i0 = 0 case, where the paper says A = {0,1,h+1,(h+1)^2} is a B4-set and concludes |hA| = binom(h+3,3); the conclusion actually requires the stronger B_h property, which is not stated or proved there. This is a genuine proof gap but not a circularity: the missing B_h property is an elementary, external fact about this specific set, and it is not derived from the theorem being proved or from any self-citation. There is also a minor typo in the h = 1 paragraph, where i0 is written as 1 rather than 0, but the listed set is correct for i0 = 0. These issues are repairable and do not make the construction depend on its own conclusion. Accordingly, the paper's derivation chain is self-contained and receives a circularity score of 0.
Assumptions & free parameters
assumptions (2)
- domain assumption The set B={0,1,h+1} is a B_m-set for every m <= h, so |mB| = binom(m+2,2).
- domain assumption The set {0,1,h+1,(h+1)^2} is a B_h-set, hence has |hA| = binom(h+3,3).
Cite this review
Pith. "Pith review of Additive sumset sizes with tetrahedral differences." pith.science (2026). https://pith.science/paper/2F6WACSB
@misc{pith2026250708646,
author = {Pith},
title = {Pith review of: Additive sumset sizes with tetrahedral differences},
year = {2026},
howpublished = {\url{https://pith.science/paper/2F6WACSB}},
note = {Machine review of arXiv:2507.08646}
}
abstract
Experimental calculations suggest that the $h$-fold sumset sizes of 4-element sets of integers are concentrated at $h$ numbers that are differences of tetrahedral numbers. In this paper it is proved that these "popular" sumset sizes always exist. Explicit $h$-adically defined sets are constructed for each of these numbers.
Forward citations
Cited by 2 Pith papers
-
Possible Sizes of Sumsets
For fixed h and large k, the possible sizes of h-fold sumsets of k-element integer sets form the full interval [hk−h+1, C(h+k−1,h)] minus C(h−1,2) specified numbers; the h=3 case is settled for all k>2.
-
On the size of $h$-fold sumsets
For A = {0,1,...,s,a,b} with b = qa + r, the compact binomial formula for the h-fold sumset size |hA| holds for all h exactly when r = 0 or qs + r ≥ a.
Reference graph
Works this paper leans on
-
[1]
Reprinted by Chelsea Publishing Company, New York, 1971
Leonard Eugene Dickson, History of the Theory of Numbers , Volume II, Carnegie Institute of Washington, 1920. Reprinted by Chelsea Publishing Company, New York, 1971
work page 1920
-
[2]
Erd˝ os and E
P. Erd˝ os and E. Szemer´ edi, On sums and products of integers, in: Studies in Pure Mathe- matics, Birkh¨ auser, Basel-Boston, 1983, pp. 213–218
1983
-
[3]
J. Fox, N. Kravitz, and S. Zhang, Finer control on relative sizes of iterated susmets, preprint, 2025
2025
-
[4]
Hegyv´ ari, On representation problems in the additive number theory, Acta Math
N. Hegyv´ ari, On representation problems in the additive number theory, Acta Math. Hungar. 72 (1996), 35–44
1996
-
[5]
Kravitz, Relative sizes of iterated sumsets, J
N. Kravitz, Relative sizes of iterated sumsets, J. Number Theory 272 (2025), 113–128,
2025
-
[6]
M. B. Nathanson, Additive Number Theory: Inverse Problems and the Geometry of Sumsets , Springer, New York, 1996
work page 1996
-
[8]
M. B. Nathanson, Problems in additive number theory, VI: Sizes of sumsets of finite sets, Acta Math. Hungar., to appear. arXiv:2411.02365, 2024
arXiv 2024
-
[9]
M. B. Nathanson, Explicit sumset sizes in additive number theory, arXiv: 2505.05329, 2025
arXiv 2025
Show all 14 references
-
[10]
M. B. Nathanson, Compression and complexity for sumset sizes in additive number theory, arXiv:2505.20998
-
[11]
M. B. Nathanson, Triangular and tetrahedral number differences of sumset sizes in additive number theory, arXiv:2506.15015
-
[12]
O’Bryant, A complete annotated bibliography of work related to Sidon sequences, Electron
K. O’Bryant, A complete annotated bibliography of work related to Sidon sequences, Electron. J. Combin. DS11 (2004), 39
2004
-
[13]
O’Bryant, On Nathanson’s triangular number phenomenon, arXiv:2506.20836
K. O’Bryant, On Nathanson’s triangular number phenomenon, arXiv:2506.20836. 8 MEL VYN B. NATHANSON
-
[14]
P´ eringuey and A
P. P´ eringuey and A. de Roton, A note on iterated sumsets races, arXiv:2505.11233
-
[15]
Schinina, On the sumset of sets of size k, arXiv:2505.07679
V. Schinina, On the sumset of sets of size k, arXiv:2505.07679. Department of Mathematics, Lehman College (CUNY), Bronx, NY 10468 Email address : melvyn.nathanson@lehman.cuny.edu
Reviewed August 6, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.