Reused writeup on Math-987 (gpt pro 5.2)

Erdős Problem 932 --- collection record

PROVENANCE AND STATUS

Official problem: https://www.erdosproblems.com/932

Saved collaborative-database status: open. Snapshot checked: 2026-09-23T17:22:24Z. Latest status and formalization links: https://teorth.github.io/erdosproblems/

This is an attributed collection of existing writeups. The model-folder names below identify their original attribution; placement in GPT_6_Astra_Ultra does not attribute their mathematical work to that model. All locally available source versions selected for this problem are reproduced below, without editing their text.

Proof claims, computations, literature statements, and completion estimates in the reused text have not been independently verified in this collection pass. A source claiming a solution does not change the saved upstream status.

Primary source (latest numbered version in the preferred model folder): attacks/erdos/gpt_pro_5.2/932.tex.

REUSED SOURCE 1

Original attribution: gpt_pro_5.2. Original file: attacks/erdos/gpt_pro_5.2/932.tex.

1) FORMAL RESTATEMENT

Let $p_r$ denote the $r$th prime, and write $g_r:=p_{r+1}-p_r$ for the prime gap. For a positive integer $n$, let $P^+(n)$ be the largest prime factor of $n$. The property in the problem is: \[ \exists\ \text{at least two integers }n\ \text{with }p_r<n<p_{r+1}\ \text{and }P^+(n)<g_r. \] The question is whether this happens for infinitely many $r$.

Edge cases: if $g_r=2$ then there are no integers strictly between $p_r$ and $p_{r+1}$; if $g_r=3$ there is exactly one integer in between, so the property cannot hold. Thus any $r$ witnessing the property must have $g_r\ge 4$.

2) QUICK LITERATURE/CONTEXT CHECK

From the provided problem statement: Erdős believed the answer is "yes" but that such $r$ are sparse, and he could show that the density of $r$ for which at least one such $n$ exists is $0$. The statement was also formalised in Lean in the Google DeepMind Formal Conjectures project (as noted in the source file).

3) ATTACK PLAN

• Try to build explicit $r$: a sufficient condition is to find a prime gap in which there are two numbers composed only of primes $<g_r$.
• Local sufficient conditions: if an integer $n$ in the gap is small compared to $g_r$, then having a prime factor $\ge g_r$ is impossible.
• Computation: scan prime gaps up to some bound and count how many $n$ in each gap satisfy $P^+(n)<g_r$.

4) WORK

Lemma 932.1 (a size condition that forces $g_r$-smoothness). Fix $r$ and set $g=g_r$. If $n$ is a composite integer satisfying $p_r<n<p_{r+1}$ and $n<2g$, then every prime factor of $n$ is $<g$. (Equivalently, $P^+(n)<g$.)

Proof. Assume $n$ is composite and $n<2g$. Let $q$ be any prime factor of $n$. If $q\ge g$, then $n=q\cdot m$ for some integer $m\ge 2$ (since $n$ is composite). Hence $n\ge 2q\ge 2g$, contradicting $n<2g$. Therefore every prime factor $q$ of $n$ satisfies $q<g$, as claimed. $\square$

Proposition 932.2 (explicit small example). For $r=4$ we have $(p_r,p_{r+1})=(7,11)$ and $g_4=4$. The integers $8$ and $9$ satisfy $7<8<9<11$ and all their prime factors are $<4$. Hence $r=4$ satisfies the property.

Proof. Compute: $8=2^3$ has prime factor set $\{2\}$, and $9=3^2$ has prime factor set $\{3\}$. Both $2$ and $3$ are strictly less than $g_4=4$. $\square$

FAST REALITY CHECK (local computation up to $200{,}000$). I computed all prime gaps with $p_{r+1}\le 200{,}000$ and counted, for each $r$, how many integers $n$ with $p_r<n<p_{r+1}$ satisfy $P^+(n)<g_r$. In this range I found $20$ values of $r$ for which there are at least two such integers. The first few are:

(Here "good" means all prime factors $<g_r$.) This is only evidence and does not address infinitude.

5) VERIFICATION

• Lemma 932.1: the only use of compositeness is to ensure $m\ge 2$ when writing $n=q m$. If $n$ were prime, it could have $P^+(n)=n\ge g$.
• Proposition 932.2: verified by direct factorization.

6) FINAL

UNRESOLVED

(i) Strongest proved partial result here. A sufficient condition for an integer $n$ in the prime gap to satisfy $P^+(n)<g_r$ is $n<2g_r$ together with compositeness (Lemma 932.1). There are explicit small $r$ with at least two such $n$ (e.g. $r=4$) and computationally many more in small ranges.

(ii) First gap (crisp). Prove that there are infinitely many $r$ such that at least two integers $n$ with $p_r<n<p_{r+1}$ satisfy $P^+(n)<p_{r+1}-p_r$.

(iii) Top 3 next moves.

1. Try to explicitly construct prime gaps that contain two numbers whose prime factors are all below the gap size (e.g. by combining known "large gap" constructions with smooth numbers).
2. Compute further to much larger bounds to guess the growth rate (frequency) of such $r$, and to look for repeating structural patterns in the good integers.
3. Seek a theoretical sufficient condition on the residue class structure of $p_r$ modulo small primes that guarantees two $g_r$-smooth integers in $(p_r,p_{r+1})$.

(iv) What a minimal counterexample would look like. A counterexample to the infinitude statement would mean that beyond some $r_0$, every prime gap $(p_r,p_{r+1})$ contains at most one integer whose prime factors are all $<g_r$. In particular, for large $r$ the interval would force most composites to have a prime factor at least the gap size.

COMPLETION ESTIMATE

The primary source supplies no extractable numerical completion estimate. Its mathematical progress is not numerically assessed here.

Newly verified progress in this collection pass: $0\%$. This measures new verification work, not the amount of mathematics achieved in the reused text.