REVIEW 5 major objections 3 minor 12 references
New Polynomial Identities and Some Consequences
T0 review · 5 major / 3 minor · reviewed 2026-08-07 · deepseek-v4-flash
Pith's one-line read Two new polynomial identities, proved by Beta-function integration, put Frisch's and Klamkin's binomial sum identities on a common footing.
desk verdict A useful toolbox paper with a genuine but fixable gap: the Beta-integral proofs require stronger convergence hypotheses than the theorems state. 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 load-bearing objects are the two polynomial identities (1.1) and (1.2), each an equality between a finite sum of $x^k$ times an inverted binomial coefficient and a second finite sum with inverted binomial coefficients times $(1+x)^{n-k}$ or $(1-x)^{n-k}$. They are proved by multiplying the binomial expansion of $(1+xy)^n$ by $y^{r-s}(1-y)^{s-1}$ or $y^s(1-y)^{r-n-s}$ and integrating over $(0,1)$, turning each term into an Euler Beta integral. The general scheme then reads combinatorial identities off any polynomial identity of the form $\sum f(k)x^{p(k)} = \sum g(k)(1-x)^{q(k)}$, and a differentiation step with $x=e^t$ converts integer moments of the coefficient sequences into Stirling-number evaluations.
What would settle it
Evaluate (1.2) at $n=1$, $r=0$, $s=-1/2$: the $k=1$ term has denominator $r-k+1=0$, so the displayed formula is undefined; checking whether the two sides approach a common limit as $s\to -1/2$ would settle whether a limiting interpretation is required. A direct failure of the published proof conditions can be seen with $n=1$, $r=1$, $s=-20.5$, where $\Re(r-n-s+1)>0$ holds but the integral in (2.4) behaves like $y^{-19.5}$ near $y=0$ and diverges.
Extended reading notes
Core claim
The paper's central claim is that identities (1.1) and (1.2) hold as genuine polynomial identities for complex $r$ and $s$ under the stated half-plane conditions, with $n$ a non-negative integer and $x$ complex. Identity (1.1) is presented as a new generalization of a known identity in the cited table, with Frisch's identity recovered at $x=-1$; identity (1.2) yields Klamkin's identity at $x=1$. From these anchors the paper derives generalized Frisch and Klamkin identities carrying an extra binomial factor, then uses derivative evaluations to obtain closed forms for sums of $k^m$ times an inverted binomial coefficient. The same scheme applied to MacMahon's identity produces formulas for $\sum_{k=0}^{2n}(-1)^k k^m \binom{2n}{k}^3$, including complements of Dixon's identity.
Load-bearing premise
The proof depends on multiplying the binomial expansion by the Beta kernel and integrating term by term on $(0,1)$, which is legitimate only if every resulting integral converges; the stated half-plane conditions do not by themselves ensure that convergence for all $k$.
Editorial extensions
If this is right
- Frisch's identity and Klamkin's identity emerge as the specializations $x=-1$ and $x=1$, showing that two classical results are faces of a single equality.
- The generalized identities evaluate sums with an extra factor $\binom{u}{n-k}$ or $\binom{u+n-k}{n-k}$, reducing to the classical identities at $u=0$ or $u=-1$.
- The moment formulas give finite Stirling-number evaluations for $\sum (-1)^k k^m \binom{n}{k}\binom{k+r}{s}^{-1}$ and $\sum k^m \binom{n}{k}\binom{r}{k+s}^{-1}$ for every non-negative integer $m$.
- Complements of Dixon's identity are obtained as the $m=1$ and $m=2$ cases of a general evaluation of $\sum_{k=0}^{2n}(-1)^k k^m \binom{2n}{k}^3$.
Reading between the lines
- Our inference: as written, the proof needs additional convergence conditions such as $\Re(k+s)>0$ and $\Re(n-k+s+1)>0$ for every $k$, and identities with zero denominators such as $r-k+1=0$ require a limiting interpretation; the final identities may still hold by analytic continuation, but that is not established in the paper.
- Our inference: since all sums are finite and the integration kernel is the standard Beta weight, the same construction could be run with other orthogonal-polynomial or $q$-weight kernels to produce $q$-analogues of Frisch- and Klamkin-type identities.
- Our inference: the Theorem 7 scheme is an automatic moment-generating machine: given any polynomial identity of the prescribed shape, differentiating the substitution $x=e^t$ yields closed evaluations of moments of its coefficient sequences, so the paper's list of applications is not exhaustive.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. Using the Euler Beta function and the binomial theorem, the paper claims two new polynomial identities, (1.1) and (1.2), for complex parameters r and s and nonnegative integer n, and from them derives Frisch's identity, Klamkin's identity, generalizations of both, Stirling-weighted variants, and complements of Dixon's identity. It also proposes a general scheme, Theorem 7, for converting polynomial identities of a certain form into summatory identities.
Significance. The paper's contribution is a unified elementary route to several known and new combinatorial identities, with Frisch's and Klamkin's identities correctly identified as specializations of the two central polynomial identities. The identities are not assumed: explicit Beta-integral proofs are attempted, and the intended connections to Gould, MacMahon, and Dixon are clear. If the proof infrastructure is repaired, the framework would be a useful toolbox for deriving such identities. However, in its present form the central derivations rely on unstated integrability conditions, at least one displayed evaluation in Lemma 1 is incorrect as printed, and the stated parameter ranges are not justified.
major comments (5)
- [§2, Lemma 1, Eq. (2.4)] As printed, Eq. (2.4) lacks the reciprocal on the binomial coefficient. The correct evaluation is \(\int_0^1 y^{n-k+s}(1-y)^{r-n-s}\,dy = (r-k+1)^{-1} \binom{r-k}{r-s-n}^{-1}\), which follows by comparing the integral with the Beta function \(B(n-k+s+1, r-n-s+1)\). With the printed display, the lemma is false, and the proof of Theorem 2 does not produce identity (1.2); for example, n=1, k=1, r=0.1, s=-0.5 gives the wrong numerical value on the printed right-hand side. Please correct this display and re-verify all later applications of (2.4), especially in §3.2 and §4.2.
- [§2, Lemma 1 and Theorems 1–2] The convergence hypotheses are incomplete. Formula (2.1) needs Re(s)>0, not merely that s is not a non-positive integer; formula (2.2) needs Re(k+s)>0 for every k; formulas (2.3) and (2.4) need Re(k+s+1)>0 and Re(n-k+s+1)>0 for every k. Theorem 1 assumes only Re(r-s+1)>0 and s not a non-positive integer, and Theorem 2 assumes only Re(r-n-s+1)>0 and s not a negative integer. The example n=1, r=1, s=-20.5 satisfies both theorems' stated assumptions, but the integrals in (2.2) and (2.4) are not in L^1(0,1). Thus the termwise Beta integration is not justified on the stated parameter ranges. The identities may be salvageable by analytic continuation, but this is neither stated nor proved.
- [§2, Eq. (1.2) and Theorem 2] The right-hand side of (1.2) divides by r-k+1, and the hypotheses do not exclude r=k-1 for some k in {0,...,n}. For n=1, r=0, s=-1/2, the k=1 term is an indeterminate 0/0. In this example the two-sided limit as r approaches 0 is finite and agrees with the left-hand side, so the singularity may be removable, but this requires an explicit statement and proof, or such r must be excluded. Since (1.2) is one of the two central identities, the parameter range of the theorem is not accurately described as printed.
- [§3, Theorems 3 and 5] The general integration scheme has the same convergence gap. In Theorem 3, the proof uses (2.1) and (2.2), so it additionally requires Re(s)>0 and Re(p(k)+s)>0 for all k; the stated conditions Re(r+min p -s+1)>0 and Re(min q + s)>0 do not imply Re(s)>0. In Theorem 5, the proof requires Re(q(k)+s+1)>0 for all k, which is not implied by the stated max conditions. Because (3.6), (3.7), (3.18), and (3.19) are the basis for the generalized Frisch and Klamkin identities, these results inherit the incomplete parameter ranges.
- [§5, Propositions 10 and 11] The same missing-integrability issue appears in later propositions. In Proposition 10, the value s=-1/2 is not a non-positive integer and is therefore allowed, but the proof integrates (1-y)^{s-2}, which is not integrable on (0,1). Proposition 11 similarly allows parameters with Re(s)<-1. If these identities are intended to hold by meromorphic continuation rather than by the displayed Beta integrals, the hypotheses and proofs should say so explicitly.
minor comments (3)
- [§4.1, Eqs. (4.4)–(4.5)] The displayed formulas for the m=1 and m=2 cases are malformed; the fractions involving s-r-1 and n+s-1 should be typeset and checked.
- [§5.3, Proof of Proposition 13] The text 'Uef(k)' should read 'Use f(k)'.
- [§4, Proof of Theorem 7] The text 'To drive (4.14)' should read 'To derive (4.14)'.
Circularity Check
No circularity: the core identities are proven from the Beta function and binomial theorem, and all later identities reduce to them one-way.
full rationale
The paper's chain is not circular. The central polynomial identities (1.1) and (1.2) are derived in Theorems 1 and 2 directly from the binomial expansion of plain algebraic identities and termwise Beta-function integration (Lemma 1); they are not assumed, fitted, or imported from prior work. Every later result, including the Frisch and Klamkin specializations (1.3)/(1.4), the generalized Frisch/Klamkin identities (Theorems 4 and 6), the Stirling-weighted extensions (Propositions 8 and 9), and the Dixon-related sums (Proposition 15), is obtained from these identities or from independently cited external identities (Gould, Simons, MacMahon, Waring) by integration or differentiation schemes. Those external inputs do not contain the paper's target identities, so the dependence is one-way. The only self-citation, Adegoke and Frontczak [2], appears in the introduction as a historical note about a previous use of Frisch's identity and is not load-bearing. The reviewer-identified gaps in the stated convergence hypotheses for the Beta integrals are genuine mathematical/correctness concerns, but they do not constitute circularity: the identities may be valid under stronger hypotheses or by analytic continuation, and the derivation method remains independent of the conclusions. Overall, the derivation is self-contained against the cited benchmark results, and no reduction of a prediction to its input by construction occurs.
Assumptions & free parameters
assumptions (4)
- standard math Euler Beta integral evaluation Gamma(a)Gamma(b)/Gamma(a+b) and its analytic continuation to complex parameters
- standard math Binomial theorem for finite sums with complex variable x
- domain assumption Known polynomial identities of Gould, Simons, and MacMahon quoted as lemmas
- standard math Stirling and r-Stirling number derivative evaluations in Lemma 4
Cite this review
Pith. "Pith review of New Polynomial Identities and Some Consequences." pith.science (2026). https://pith.science/paper/YDAQK6HW
@misc{pith2026250606617,
author = {Pith},
title = {Pith review of: New Polynomial Identities and Some Consequences},
year = {2026},
howpublished = {\url{https://pith.science/paper/YDAQK6HW}},
note = {Machine review of arXiv:2506.06617}
}
read the original abstract
Using an elementary approach involving the Euler Beta function and the binomial theorem, we derive two polynomial identities; one of which is a generalization of a known polynomial identity. Two well-known combinatorial identities, namely Frisch's identity and Klamkin's identity, appear as immediate consequences of the polynomial identities. We subsequently establish several combinatorial identities, including a generalization of each of Frisch's identity and Klamkin's identity. Finally, we develop a scheme for deriving combinatorial identities associated with polynomial identities of a certain type.
Reference graph
Works this paper leans on
-
[1]
Abel, A short proof of the binomial identities of Frisch and Klamkin, J
U. Abel, A short proof of the binomial identities of Frisch and Klamkin, J. Integer Seq. 23 (2020), Article 20.7.1
work page 2020
-
[2]
K. Adegoke and R. Frontczak, Some notes on an identity of Frisch,Open J. Math. Sci. 8(2024), 216–226. 18
work page 2024
-
[3]
A. C. Dixon, Summation of a certain series,Proc. London Math. Soc.35(1903), 284– 289
work page 1903
-
[4]
R. Frisch, Sur les semi-invariants et moments employ´ es dans l´ etude des distributions statistiques,Skrifter utgitt av Det Norske Videnskaps-Akademi i Oslo, II, Historisk- Filosofisk Klasse,3(1926), 1–87
work page 1926
-
[5]
H. W. Gould,Combinatorial Identities, Morgantown, West Virginia, 1972
work page 1972
-
[6]
H. W. Gould, The Girard-Waring power sum formulas for symmetric functions and Fibonacci sequences,Fibonacci Quart.37(1999), 135–140
work page 1999
-
[7]
H. W. Gould and J. Quaintance, On the binomial identities of Frisch and Klamkin, J. Integer Seq. 19 (2016), Article 16.7.7
work page 2016
-
[8]
D. Laissaoui and M. Rahmani, An explicit formula for sums of powers of integers in terms of Stirling numbers,J. Integer Seq.20(2017), Article 17.4.8
work page 2017
Show all 12 references
-
[9]
A. M. Rockett, Sums of the inverses of binomial coefficients,Fibonacci Quart.19(1981), 433–437
1981
-
[10]
Simons, A curious identity,Math
S. Simons, A curious identity,Math. Gaz.85(2001), 296–298
2001
-
[11]
Sun, On sums involving products of three binomial coefficients,Acta Arith.156 (2012), 123–141
Z. Sun, On sums involving products of three binomial coefficients,Acta Arith.156 (2012), 123–141
2012
-
[12]
Ward, 100 years of Dixon’s identity,IMS Bulletin27(1991), 46–54
J. Ward, 100 years of Dixon’s identity,IMS Bulletin27(1991), 46–54. 19
1991
Reviewed August 7, 2026 · model on record in the stance chip above.
Discussion (0). Sign in to comment.