On Waring-Goldbach problem for two squares and higher powers

arXiv:2610.15142 · math.NT · Submitted 2026-10-10 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Mathematics. The day in mathematics: new theorems, new proofs, and what they open.

Ada: We're looking at the paper: "On Waring-Goldbach problem for two squares and higher powers".

Emmy: Here's the gist of the paper. Every sufficiently large odd integer n can be represented as n = p2 1 + p2 squared + p4 cubed + p4 4 + p5 5 + p6 7, where p1,..., p7 are prime variables n=p 2 1+p 2 2+p 4 3+p 4 4+p 5 5+p 6 7, where p1,..., p7 are prime variables Theorem 1.1 states that every sufficiently large odd integer n can be represented as n = p2 1 + p2 squared + p4 cubed + p4 4 + p5 5 + p6 7, where p1,..., p7 are primes This result is an enhancement upon a previous result of Andrade

1: The proof proceeds by establishing bounds for the integral ν(n) = Z 0 to F(α)e(-αn) dα (2.4), aiming to show that the lower bound ν(n) ≫ nΘ (2.6) holds for all sufficiently large odd integers n This is a quantitative form of the theorem The goal is to demonstrate that the lower bound ν(n) ≫ nΘ (2.6) holds for all sufficiently large odd integers n This requires establishing two bounds: Z M F(α)e(-αn) dα ≫ nΘ and Z m F(α) dα ≪ nΘL−1 (2.7) for every sufficiently large odd integer n Major Arcs Asymptotics For k = 2, 4, 5, 7, the authors write Sk(q, a) = X 1 <= x <= q (x,q)=1 e(axk/q), vk(β) = 1/k X 2−kn<m<=n m1/k−1 e(βm) (k = 2, 4, 7) Lemma 3.1 provides the result that for α ∈ M(q, a), fk(α) = Sk(q, a)/ϕ(q) vk(α − a/q) + O Pke−c1√L Lemma 3.2 proves Z M F(α)e(-αn) dα ≫ nΘ This involves approximating f5(α) and multiplying the approximations for fk(α) to estimate F(α), leading to (3.2) Minor Arcs Estimate The contribution from the minor arcs is bounded by Z m F(α) dα ≪ nΘL−1 (4.13) This bound is derived by combining Lemma 4.5 and (4.13) The estimate Z E F(α) dα ≪ nΘ−η (4.7) is established by applying Hölder’s inequality to infer that Z E F(α) dα ⩽ J 1/4 1 J 3/4 sup α∈E f7(α) 1/2 The complement of E is treated using Lemma 4.2, leading to the conclusion that

0, 1: - E ⊆ R (4.9) Combining the bounds in (2.7) with Lemma 3.2 and (4.14), it is known that the two bounds in (2.7) hold This completes the proof of Theorem 1.1 The final step involves showing that S(n) ≫ 1 for all large odd n, which is achieved by proving that the integral on the right–hand side of (3.7) is ≫ nΘ−θ/5 This integral yields Z LB/n − LB/n w(β)e(-βn) dβ = Z 1/2−1/2 w(β)e(-βn) dβ + O nΘ−θ/5L−B The integral on the right–hand side of (3.7) is ≫ nΘ−θ/5, which combined with Pθ/5 = nθ/5, (3.2), (3.3) and the property S(n) ≫ 1 completes the proof of Lemma 3.

Title and authors: Ada: So we're looking at this paper titled "On Waring-Goldbach problem for two squares and higher powers." It tackles how we can represent large odd integers as sums of different prime powers.

Emmy: It focuses on finding a representation for every sufficiently large odd integer n. Specifically, the main result is that n can be written as the sum of six terms: one square of a prime, another square of a prime, and then three different higher powers involving primes.

Kolya: That's interesting because it builds on older results. I mean we already had some work showing that certain integers could be written as sums of squares of primes or biquadrates, but this paper seems to push the combination further than before.

Ada: Exactly. The specific representation they prove is n equals one prime squared, another prime squared, and then three fourth powers of primes, a fifth power of a prime, and a seventh power of a prime. That’s what the authors call theorem one point one in the paper > <ref:2610.15142#pg1>

Emmy: And that's quite specific because they are replacing some variables from previous work with different ones to get this stronger result for large odd integers >

Kolya: What makes it significant, though, is that they aren't just proving existence; they are doing a quantitative analysis. They are aiming to show a lower bound on the number of solutions for this equation >

Ada: Right. They use an integral called the nu function to count those representations, and their main goal is to prove that this count gets very large compared to n itself when n is big enough >

Emmy: They set up this analysis using the circle method, which splits the problem into major arcs and minor arcs. They want to show that the contribution from those major arcs dominates completely >

Kolya: And for that, they need two key things to happen simultaneously. First, they need a strong lower bound on the main arc integral that grows with n >

Ada: Right, so the first thing is showing that this main arc integral grows faster than n itself when you look at large odd integers >

Emmy: And the second thing is bounding the contribution from all those minor arcs. They need to show that those minor arc contributions don't overwhelm the main part of the count >

Kolya: So, they set up these two bounds together, which they call two point seven in their text, and they show that both of those conditions are met for sufficiently large odd integers >

Ada: And then they use some estimates from previous work to prove that these two bounds actually hold, which completes theorem one point one > <ref:2610.15142#pg1>

Emmy: It’s a lot of machinery to get there, but the structure of how they handle the major arcs versus the minor arcs is pretty elegant, showing how it works in practice >

Kolya: It's definitely quite technical. I wonder if this type of quantitative bound on the density of these representations has broader implications for other additive problems in number theory >

Ada: Definitely. If we can prove that every large odd integer has this kind of representation, it suggests a deeper structure to how primes distribute themselves across different powers >

Emmy: And it connects back to the general philosophy of the Hardy–Littlewood method, which is about finding asymptotic formulas for sums involving primes >

Kolya: So, when we look at this paper "On Waring-Goldbach problem for two squares and higher powers," we see a refinement of an existing result using more complex prime powers >

Ada: Right. And the key part is that they prove a specific formula involving these different prime powers holds for all sufficiently large odd integers >

Emmy: So, to summarize, the paper establishes a precise representation of every large odd integer as that sum of six specific prime powers and provides a quantitative proof showing this happens with high frequency >

Kolya: It’s about confirming what we suspected based on older conjectures by providing the necessary analytic machinery to make it rigorous for all sufficiently large n >

The paper's summary: Ada: So this paper, "On Waring-Goldbach problem for two squares and higher powers," it's about taking these large odd integers we know exist and showing they can be written as a sum of six specific prime powers.

Emmy: Exactly what you said Ada, it’s not just about *if* those numbers can be represented, but proving that *every* sufficiently large odd integer has this representation >

Ada: Right. They’re using a very specific combination of terms here—one square of a prime, another square of a prime, and then three different higher powers involving primes like fourth power, fifth power, and seventh power >

Emmy: And the structure they’re using to prove it is based on the circle method. They split the problem into two parts: what happens on those main arcs where things are dense, and what happens on those minor arcs where things are sparse >

Ada: It's about showing that the main arc part of the count is big enough to dominate everything else when you look at really large odd numbers >

Emmy: And then they have to show that those minor arc contributions don't sneak in and ruin the estimate, which they do by using some clever estimates from previous work >

Ada: It's quite a tight setup. They establish two bounds that have to hold together for the theorem to work, and they manage to prove both of those bounds are true for any large odd number >

Emmy: And what’s really interesting is how they handle those minor arc estimates. They use Hölder’s inequality and some geometric arguments about the set of bad values, which lets them bound that contribution very tightly >

Ada: It means they can get a quantitative measure for how often this representation occurs, showing that it happens with a frequency that scales nicely with the size of the odd number >

Emmy: That quantitative aspect is what makes it more than just an existence proof. It gives us a way to predict how dense these representations are in the set of all large odd integers >

Ada: And for someone just listening, what this means is that we have a concrete target. We know that for any odd number big enough, this specific sum of six prime powers will always exist >

Emmy: It connects back to the general idea of additive problems involving primes. It shows that even when you mix different types of prime powers together, the structure allows for representation across all large odd integers >

Ada: So they’ve taken a problem about sums and turned it into a precise statement about density for big numbers >

Emmy: And that density estimate is what we need to keep in mind as we look at other additive problems involving different types of number theoretic objects, like the ones we're hearing about next >

The paper's improvements: Kolya: So we’ve seen how this paper establishes a quantitative lower bound for representing big odd numbers as sums of those six specific prime powers, and now Ada, what are the authors suggesting to make it even better?

Ada: Well, they’re looking at ways to refine the representation itself. They aren't just settling for *any* set of primes; they’re trying to find a tighter structure on which combinations work best >

Kolya: Tighter structure how that helps with the probabilistic view? Does it change which numbers are representable?

Ada: It changes the probability, Kolya. They’re exploring ideas like iterative exceptional set estimates to figure out when a number can be written as that sum, but now they’re adding conditions on how many prime factors the first term can have >

Emmy: That sounds like moving from just counting solutions to analyzing the structure of those solutions themselves. It’s trying to impose more algebraic constraints on the primes involved >

Kolya: So if you tighten the constraint on how many prime factors a certain term has, does that make the bound stronger? Is it still a big number?

Ada: The goal is to get a representation where the first term, for instance, has at most four prime factors. They’re trying to show that even with this restriction on the complexity of one part of the sum, we can still guarantee coverage for all sufficiently large odd numbers >

Emmy: It’s a structural improvement. Instead of just saying "there is a solution," they are proving that within this restricted set of solutions, there's enough density to cover everything >

Kolya: So it’s less about finding *a* representation and more about characterizing the *type* of representations that are guaranteed to exist in a dense way. That’s interesting for understanding the distribution >

Ada: Exactly. It moves from a broad statement about existence to a more granular statement about the density of those specific types of sums, which is what we need for number theory >

Emmy: And this leads us into how these constraints interact with other systems. They’re looking at how these structural limitations play out in relation to other additive problems we’ve been studying >

Kolya: It sounds like they are setting up a framework where you can look at the tail behavior of the representation count more precisely, which is what I care about most >

Conclusion: Ada: So we’ve gone through the whole thing on "On Waring-Goldbach problem for two squares and higher powers," and basically, they’ve shown that every sufficiently large odd integer has a very specific way to be written as that sum of six prime powers >

Emmy: Right. It proves existence quantitatively, meaning we can actually measure how often this representation happens as the numbers get bigger >

Ada: And they did it by carefully managing the main arcs and bounding those minor arcs so that the count is guaranteed to be large enough for any big odd number >

Kolya: From a probabilistic standpoint, what this tells us is that we’re not just looking at isolated cases; we have a solid tail bound on how often these solutions occur in the set of all odd integers >

Emmy: That tail bound is crucial because it shows the distribution isn't just random noise; there’s an underlying arithmetic structure governing these representations >

Ada: It confirms that even with those specific, complicated exponents like the seventh power, the density remains high enough to cover everything we expect for large odd numbers >

Kolya: I wonder if those constants they derived really survive when you try to apply this to something more complex than just odd integers, say a different set of primes?

Emmy: That's exactly what they’re hinting at. It suggests that the techniques used here, like the way they handle the integral estimates over alpha, might have broader applications in understanding other additive problems >

Ada: So it’s not just about this one equation; it’s about refining the tools we use to tackle these kinds of representation questions across different areas of number theory >

Kolya: It definitely gives me confidence that this type of quantitative analysis is a useful path forward for proving other conjectures in additive number theory >

Linji Long, Jinjiang Li, Min Zhang

China University of Mining and Technology · Beijing Information Science and Technology University

math.NT

Submitted: 2026-10-10

Updated: 2026-11-08

Comments: 12 pages

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Importance score: 65/100

The gist: Every sufficiently large odd integer n can be represented as n = p2 1 + p2 squared + p4 cubed + p4 4 + p5 5 + p6 7, where p1,..., p7 are prime variables n=p 2 1+p 2 2+p 4 3+p 4 4+p 5 5+p 6 7…

Key concepts

Integral ν(n)
This is a mathematical tool used to count the number of ways an integer n can be represented by a specific formula involving primes. The goal is to show that this integral provides a lower bound that grows faster than n itself, which is necessary to prove the main theorem.
Circle Method
This is a powerful technique used in analytic number theory to estimate the number of solutions to an equation. It involves breaking down the problem into contributions from 'major arcs' (where solutions are expected) and 'minor arcs' (where solutions are rare).
Major Arcs Asymptotics
This part of the proof focuses on approximating functions related to the representation formula over specific ranges called major arcs. It uses specialized formulas like Sk(q, a) and vk(β) to show that the main contribution to the count is large enough.
Minor Arcs Estimate
This involves bounding the contribution from 'minor arcs'—the parts of the circle method where solutions are sparse. By using Hölder's inequality and other lemmas, the authors show this error term is small enough not to invalidate the main result.

Terminology

Summary

Every sufficiently large odd integer n can be represented as n = p2 1 + p2 squared + p4 cubed + p4 4 + p5 5 + p6 7, where p1,. p. 0.., p7 are prime variables n=p 2 1+p 2 2+p 4 3+p 4 4+p 5 5+p 6 7, where p1,. p. 0.., p7 are prime variables

Main Result

Theorem 1.1 states that every sufficiently large odd integer n can be represented as n = p2 1 + p2 squared + p4 cubed + p4 4 + p5 5 + p6 7, where p1,. p. 0.., p7 are primes This result is an enhancement upon a previous result of Andrade [1]

Methodology Outline

The proof proceeds by establishing bounds for the integral ν(n) = Z 0 to F(α)e(-αn) dα (2.4), aiming to show that the lower bound ν(n) ≫ nΘ (2. p. 46) holds for all sufficiently large odd integers n This is a quantitative form of the theorem The goal is to demonstrate that the lower bound ν(n) ≫ nΘ (2. p. 16) holds for all sufficiently large odd integers n This requires establishing two bounds: Z M F(α)e(-αn) dα ≫ nΘ and Z m F(α) dα ≪ nΘL−1 (2. p. 17) for every sufficiently large odd integer n

Major Arcs Asymptotics

For k = 2, 4, 5, 7, the authors write Sk(q, a) = X 1 <= x <= q (x,q)=1 e(axk/q), vk(β) = 1/k X 2−kn<m<=n m1/k−1 e(βm) (k = 2, 4, 7) Lemma 3.1 provides the result that for α ∈ M(q, a), fk(α) = Sk(q, a)/ϕ(q) vk(α − a/q) + O Pke−c1√L Lemma 3. p. 22 proves Z M F(α)e(-αn) dα ≫ nΘ This involves approximating f5(α) and multiplying the approximations for fk(α) to estimate F(α), leading to (3. p. 22)

Minor Arcs Estimate

The contribution from the minor arcs is bounded by Z m F(α) dα ≪ nΘL−1 (4.13) This bound is derived by combining Lemma 4. p. 75 and (4.13) The estimate Z E F(α) dα ≪ nΘ−η (4. p. 27) is established by applying Hölder’s inequality to infer that Z E F(α) dα ⩽ J 1/4 1 J 3/4 sup α∈E f7(α) 1/2 The complement of E is treated using Lemma 4. p. 72, leading to the conclusion that [0, 1] - E ⊆ R (4. p. 69)

Conclusion

Combining the bounds in (2.7) with Lemma 3. p. 22 and (4.14), it is known that the two bounds in (2. p. 97) hold This completes the proof of Theorem 1. p. 91 The final step involves showing that S(n) ≫ 1 for all large odd n, which is achieved by proving that the integral on the right–hand side of (3. p. 47) is ≫ nΘ−θ/5 This integral yields Z LB/n − LB/n w(β)e(-βn) dβ = Z 1/2−1/2 w(β)e(-βn) dβ + O nΘ−θ/5L−B The integral on the right–hand side of (3. p. 47) is ≫ nΘ−θ/5, which combined with Pθ/5 = nθ/5, (3. p. 52), (3.3) and the property S(n) ≫ 1 completes the proof of Lemma 3. p. 5

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Enhanced representation of large odd integers through a specific sum of prime powers: The system can prove that every sufficiently large odd integer n can be represented as n = p 2 1 + p 2 2 + p 4 3 + p 4 4 + p 4 5 + p 5 6 + p 7 where primes are the variables.

  2. Quantitative verification of additive representation bounds: The improved system can establish a quantitative form of the theorem by demonstrating that the lower bound ν(n) ≫ n Θ (2.6) holds for all sufficiently large odd integers n.

  3. Refined circle method application for sparse representations: The AI can utilize iterative exceptional set estimates to prove the main conjecture (1.1), allowing it to determine when a number can be written as n = x squared + p 2 2 + p 4 3 + p 4 4 + p 5 3 + p 6 4, where x has at most four prime factors.

  4. Precise analysis of local conditions modulo small primes: The system can perform detailed modular arithmetic checks to show that every residue class modulo 3 is represented and that the last two variables alone can supply any prescribed residue modulo 5, based on the properties of the terms in equation (3.6).

  5. Robust estimation of minor arc contributions: The AI can bound the contribution from minor arcs by showing Z m F(α) dα ≪ nΘ−η + nΘL−1, which allows for a rigorous proof that the two bounds in (2.7) hold and completes the proof of Theorem 1.1.

Related papers