🔢 mathematics

Prime Certificates for Exact Vertex-Coprime Ramsey Numbers

This paper establishes exact formulas for mixed vertex- and edge-coloring coprime Ramsey numbers on the coprime graph by utilizing elementary prime-based certificates, specifically proving that the vertex-coloring number equals the $m$ -th prime where $m$ is the sum of the clique sizes minus one, and that the edge-coloring number reduces to a classical Ramsey number via a prime-index transfer.

Original authors: Zhicheng Du, Wenji Xi, Zhuo Deng, Lan Ma

Published 2026-05-27

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Zhicheng Du, Wenji Xi, Zhuo Deng, Lan Ma

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you have a giant room filled with people, numbered from 1 to $n$ . In this room, two people are considered "friends" if their numbers share no common factors other than 1 (mathematicians call this being "coprime"). For example, 3 and 4 are friends, but 4 and 6 are not (they both share a factor of 2).

This paper solves a puzzle about how to color these people with different colored shirts (say, Red, Blue, Green, etc.) without creating a specific "forbidden" pattern. The forbidden pattern is a group of friends who all wear the same shirt.

The Big Question

The authors ask: How big does the room need to be ( $n$ ) before you are forced to have a group of $k$ mutual friends all wearing the same color?

In the world of standard math puzzles (called Ramsey Theory), the answer is usually a huge, messy number that is incredibly hard to calculate. You often have to run supercomputers to guess the answer for even small groups.

The Surprise Discovery

The authors found that for this specific "coprime" room, the answer is surprisingly simple and exact. It depends entirely on prime numbers (numbers like 2, 3, 5, 7, 11... that can't be divided evenly by anything else).

The formula they discovered is:

The answer is the $M$ -th prime number.

Where $M$ is calculated by adding up how many extra friends you need for each color, minus one.

If you want to avoid a group of 3 Red friends and 3 Blue friends, you calculate $(3-1) + (3-1) = 4$ .
The answer is the 4th prime number, which is 7.
This means if you have 7 people, no matter how you color them, you must have a group of 3 mutual friends in one color. If you only have 6 people, you can color them to avoid this.

How Did They Solve It? (The "Prime Bin" Analogy)

The authors didn't use a supercomputer. They used a clever "certificate" (a proof) based on two ideas:

The "Prime Clique" (The Upper Bound):
Imagine a special group of people in the room: the number 1 and all the prime numbers (2, 3, 5, 7...).
- The number 1 is friends with everyone.
- Every prime number is friends with every other prime number (because they share no factors).
- This creates a perfect "friend circle" (a clique) made entirely of primes.
- If you have enough primes in the room, the Pigeonhole Principle kicks in: if you try to put these prime-friends into colored bins, one bin must get too many of them. That bin becomes your forbidden group. This proves the answer can't be higher than a certain prime number.
The "Prime Bin" Coloring (The Lower Bound):
To prove the answer isn't lower than that prime number, they showed you can actually color the room to avoid the forbidden group.
- They took all the prime numbers and split them into "bins" (groups) corresponding to the colors.
- Every other number (the composite numbers like 4, 6, 8, 9) is colored based on one of its prime factors.
- Analogy: Imagine every composite number is a child. The child picks a "parent" (a prime factor) and wears the same shirt as that parent.
- Because the primes in each bin are limited, and every child is tied to a specific parent, you can never build a big enough group of mutual friends in any single color.

Why This Matters

It collapses a giant search: Usually, solving these problems requires checking millions of possibilities (like a SAT solver). Here, the "search" collapses into a simple check of prime numbers.
It's not random: In many math problems, the answer feels like it comes from a chaotic, random mess. Here, the structure is rigid and controlled by the "skeleton" of prime numbers.
It fixes past errors: The paper notes that previous computer attempts to solve this for a group size of 10 got the answer wrong (guessing 53). The authors proved the correct answer is 61 (the 18th prime), showing that the computer was looking at the wrong structure.

What About Other Scenarios?

The paper also looked at variations:

Edge Coloring: If you color the connections (friendships) instead of the people, the answer is still a prime number, but it's the prime number corresponding to the answer of a different, classic math puzzle. It's like a translation.
Balanced Colors: What if you demand the Red and Blue groups must be exactly the same size? Surprisingly, the answer is still the same prime number. The authors found a specific way to shuffle the "children" (composite numbers) to make the groups perfectly balanced without breaking the rules.
Moving the Room: If you start the room at number 100 instead of 1 (a "shifted interval"), the magic breaks. The simple formula no longer works because you lose the special "number 1" and the perfect start of the prime sequence. This shows the formula is very sensitive to the starting conditions.

In Summary

This paper is a detective story where the detectives realized that a chaotic-looking room of numbers actually has a very orderly secret: Prime numbers are the bosses. By understanding how primes organize the room, they found a simple, exact formula for a problem that usually requires massive computing power. They didn't just guess; they built a "prime bin" system that proves exactly where the line is drawn.

Technical Summary: Prime Certificates for Exact Vertex-Coprime Ramsey Numbers

Problem Statement
The paper investigates the mixed vertex-coprime Ramsey number, denoted $R_{cop}(k_1, \dots, k_c)$ , defined on the coprime graph $G_n$ . The vertex set of $G_n$ is $\{1, \dots, n\}$ , where an edge exists between distinct vertices $a$ and $b$ if and only if $\gcd(a, b) = 1$ . The problem asks for the smallest integer $n$ such that every $c$ -coloring of the vertices of $G_n$ contains a monochromatic clique of size $k_i$ in color $i$ for some $i \in \{1, \dots, c\}$ . Equivalently, it seeks the threshold where any partition of $\{1, \dots, n\}$ into $c$ sets must contain a set of $k_i$ pairwise coprime integers in at least one part.

Methodology and Core Mechanism
The authors resolve this problem by identifying a structural "certificate" that collapses the search space, contrasting it with the exponential complexity of direct SAT encodings. The methodology relies on two complementary constructions based on prime divisors:

Upper Bound (Prime Clique): The set $\{1\} \cup \{p \le n : p \text{ is prime}\}$ forms a clique in $G_n$ of size $\pi(n) + 1$ , where $\pi(n)$ is the prime-counting function. By the pigeonhole principle, if the total capacity of the color classes (sum of $k_i - 1$ ) is less than the size of this clique, a monochromatic clique of the required size is forced.
Lower Bound (Prime-Bin Partition): For $n$ below the threshold, the authors construct a valid coloring that avoids monochromatic cliques. This is achieved by partitioning the available primes into "bins" with capacities corresponding to $k_i - 1$ (with an adjustment for the universal vertex 1). Every composite number is colored based on a chosen prime divisor falling into one of these bins. Since pairwise coprime integers cannot share a prime divisor, the size of any monochromatic clique is bounded by the capacity of the bin containing its witness primes.

This approach is generalized to support-disjointness graphs, where vertices carry supports (subsets of atoms) and adjacency is defined by disjoint supports. The coprime graph is a specific instance where the atoms are prime numbers.

Key Contributions and Results

Exact Mixed Formula: The paper proves the exact formula for the mixed vertex-coprime Ramsey number:
$R_{cop}(k_1, \dots, k_c) = p_M$
where $M = \sum_{i=1}^c (k_i - 1)$ and $p_M$ is the $M$ -th prime number. This result holds for all $c \ge 1$ and $k_i \ge 2$ .
Edge-Coloring Reduction: The paper distinguishes vertex coloring from edge coloring on the same host graph. It establishes an exact reduction for the edge-coprime Ramsey number $R_{cop}^{edge}$ to classical complete-graph Ramsey numbers $R_{cl}$ :
$R_{cop}^{edge}(k_1, \dots, k_c) = p_{R_{cl}(k_1, \dots, k_c) - 1}$
This demonstrates that edge-coprime values are not new arithmetic constants but are classical Ramsey numbers viewed through a prime-index map.
Balanced Two-Color Threshold: The authors address the concern that the lower-bound construction might rely on highly unbalanced color classes. They prove that for the two-color diagonal case, the balanced threshold (where color class sizes differ by at most 1) is identical to the unrestricted threshold: $L_{bal}(k; 2) = p_{2k-2} - 1$ . This is achieved via a deterministic prime-bin split and a weak prime-index gap inequality ( $2p_m < p_{2m} < 3p_m$ ).
Multicolor Balance and Boundary Cases:
- For $c \ge 3$ , the balanced threshold is not exact for small $k$ but becomes exact for all sufficiently large $k$ (eventual exactness), proven via a Hall-type argument and Selberg–Delange estimates.
- The paper analyzes shifted intervals $\{m+1, \dots, m+n\}$ . Without the universal vertex 1 and the initial prime clique, the exact formula fails, and the threshold depends on local prime-factor structures.
Algorithmic Primitive: A polynomial-time certificate-extraction primitive is proposed to verify if a given graph admits a support-disjointness certificate. If verified, the Ramsey threshold and extremal colorings are determined without search.

Significance and Scope
The paper claims a complete solution for the vertex-coloring problem on the integer coprime graph, replacing large-scale computational searches (e.g., SAT encodings with millions of clauses) with a small, explicit prime-index certificate.

Structural Insight: The work highlights that the arithmetic rigidity of the coprime graph (controlled by prime divisors) allows the Ramsey problem to collapse into a simple bin-packing problem on primes, unlike the asymptotically difficult classical edge Ramsey problems on complete graphs.
Clarification of Prior Work: The authors clarify that previous computational results on "coprime Ramsey numbers" (specifically those by Towell) referred to the edge-coloring variant, which the paper shows is equivalent to classical Ramsey numbers. The vertex-coloring formula presented here was not present in checked public sources.
Limitations: The paper explicitly states it does not improve classical edge Ramsey bounds or propose a general AI-assisted solver. The results are specific to the coprime graph and its support-disjointness variants. The paper identifies open problems regarding non-complete edge targets (paths, cycles), density-constrained multicolor balance for small $k$ , and Ramsey thresholds for other arithmetic graphs where the support-disjointness property does not fully explain the adjacency.

In summary, the paper provides a definitive, elementary proof for a specific class of arithmetic Ramsey numbers, demonstrating that for this host graph, the "obstruction" is an explicit prime clique rather than a hidden random-like configuration.