Lean statement

Erdős’s reciprocal-sum arithmetic-progression conjecture

Lean is a language for mathematical statements and proofs that a computer can verify. This page shows one statement file from OpenAI’s release, possible paper matches and independent reports. The code-like names below identify the definitions or theorems selected by the proof-verification settings.

Erdős's reciprocal-sum arithmetic-progression conjecture

Theorem selected for verification: OAI.Erdos3.manuscriptReciprocalProgressionTheorem

Lean statement file: ErdosReciprocal.leanShow on the mapOpenAI’s Lean catalogue does not list this as a main result; the release’s Lean notes for family 159 link it.

See the candidate theorem Who ran the proof-verification program Report a comparison with the paper (GitHub issue)

The Lean statement

The lines of the Lean statement that state the theorem, then the definitions in the file that it uses.

def ReciprocalProgressionTheorem : Prop :=  ∀ A : Set ℕ, ¬ Summable (reciprocalTerm A) → ∀ k : ℕ, HasAP A k⋮theorem manuscriptReciprocalProgressionTheorem : ReciprocalProgressionTheorem := by  sorry

A Lean statement states the theorem and leaves its proof as sorry on purpose. OpenAI’s proof is in a separate solution module, and a program called Comparator checks that proof against this statement.

Definitions needed to read it

Local definitions the statement names, followed through the definitions they name in turn. The match is by name in the source text, so a name reached only through notation can be missed; the full file is below.

definition OAI.Erdos3.HasAP line 7

def HasAP (A : Set ℕ) (k : ℕ) : Prop :=  ∃ a d : ℕ, 0 < d ∧ ∀ i < k, a + i * d ∈ A

definition OAI.Erdos3.reciprocalTerm line 16

noncomputable def reciprocalTerm (A : Set ℕ) (n : ℕ) : ℝ := by  classical  exact if n ∈ A then (n : ℝ)⁻¹ else 0

Possible paper match, with no recorded review

No paper statement clearly matched the Lean wording. Statement Map, the program that collects Atlas’s source records, selected this possible match using words in OpenAI’s catalogue and Lean note. No review of whether the statements mean the same thing is recorded.

Shared with the catalogue’s wording or OpenAI’s note, none of them matched in the Lean text: arithmetic, contains, divergent, finite, length, progressions and reciprocal.

Corollary 1.2 of Quasipolynomial Bounds for Arithmetic Progressions

Corollary (Divergent reciprocal sums)

Every set A\subseteq\mathbb N with \sum_{a\in A}a^{-1}=\infty contains nonconstant arithmetic progressions of every finite length.

Section 1, Introduction Read it in the paper TeX source, line 32

Show the TeX

Every set $A\subseteq\mathbb N$ with
$\sum_{a\in A}a^{-1}=\infty$ contains nonconstant arithmetic
progressions of every finite length.

Other possible paper matches, closest wording first

Corollary 11.5 of Quasipolynomial Bounds for Arithmetic Progressions. Words of the Lean statement, its names split into words, that also appear in the theorem: real, natural, term, integer and progression. Shared only with the release’s name for the Lean statement or OpenAI’s note on it: free.
Corollary 11.2 of Quasipolynomial Bounds for Arithmetic Progressions. Words of the Lean statement, its names split into words, that also appear in the theorem: natural, term, integer and progression. Shared only with the release’s name for the Lean statement or OpenAI’s note on it: free and bound.
Proposition 11.1 of Quasipolynomial Bounds for Arithmetic Progressions. Words of the Lean statement, its names split into words, that also appear in the theorem: summation, natural, term, integer and progression.
Theorem 1.1 of Quasipolynomial Bounds for Arithmetic Progressions. Words of the Lean statement, its names split into words, that also appear in the theorem: term, integer and progression.
Theorem 3.4 of Quasipolynomial Bounds for Arithmetic Progressions. Words of the Lean statement, its names split into words, that also appear in the theorem: reciprocal and terms. Shared only with the release’s name for the Lean statement or OpenAI’s note on it: selected and bound.
Corollary 11.3 of Quasipolynomial Bounds for Arithmetic Progressions. Words of the Lean statement, its names split into words, that also appear in the theorem: term and progression. Shared only with the release’s name for the Lean statement or OpenAI’s note on it: free and bound.

How Atlas selects possible matches: it compares words in the Lean statement with every statement in the family. It also splits Lean names into words. The release’s label and note receive less weight; a main theorem wins a close match. If the Lean wording yields no clear match, Atlas uses the catalogue description. This selects passages to compare; it does not establish that they mean the same thing.

Reported Comparator runs

two independent groups report that Comparator accepted runs for this Lean statement file: MathVet; Neal McBurnett’s openai-math-lean-tooling. Each group reports its own Comparator run for this statement file. Atlas has not repeated the runs.

For a configured theorem, Comparator tests whether OpenAI’s proof proves the exact Lean statement. Some files instead configure definitions for comparison by name and type. Acceptance does not establish that the statement means the same thing as the paper’s theorem.

Carlos Toledo’s cert-machine

How they ran it, in their words

Every Comparator challenge re-run on Linux with nanoda switched on, decided by a preregistered rule; every family’s challenge statements read against OpenAI’s headline, clause by clause. Decider: Comparator (leanprover/comparator d03acab1 = tag v4.34.0, built with the release’s toolchain v4.34.1), lean4export 076e8e57, landrun 811cfff5, nanoda_lib 3a240721 — on GitHub’s Linux runners (.github/workflows/openai-math-kernel.yml), one j…

No verdict reached Their term: “REFUSED”.
Re-run through Comparator did not reach a verdict (the job stopped before Comparator ran (see the run log)).
Reported 10 October 2026, at the release version shown here. Original verification report

MathVet

How they ran it, in their words

An independent, model-assisted, pre-referee reading of each family’s challenge statements against its headline claim under a published rubric (PROTOCOL.md), with its own Comparator runs per challenge. Its protocol describes a planned error-rate study; this import does not establish that the study has been completed.

Accepted Their term: “comparator-pass”.
Re-run through Comparator: accepted.
Reported 8 October 2026, at the release version shown here. Original verification report

Neal McBurnett’s openai-math-lean-tooling

How they ran it, in their words

Each challenge’s solution is built and checked by Comparator at the pinned release, on Lean leanprover/lean4:v4.34.1, recording the verdict, the check time and the peak memory; the run was still in progress at this snapshot (192 accepted, 224 pending).

Accepted Their term: “accepted”.
Re-run through Comparator with Lean’s kernel: accepted (the check took 2472 s; peak memory 14.3 GiB).
Reported 10 October 2026, at the release version shown here. Original verification report

What an accepted Comparator run would establish

If Comparator is run under this configuration and accepts, the solution proves the configured theorem using only the allowed axioms, and the Lean kernel accepts every step.

This configuration lists no definition holes, so every definition the statement uses is compared as it is written in this file.

Allowed axioms: propext, Quot.sound, Classical.choice

Acceptance leaves open whether the Lean statement means the same thing as the paper’s theorem, whether the paper’s other claims are correct, and whether the result is new. Those questions require comparing the definitions and statements, reviewing the paper’s arguments, and reading the earlier literature.

Do family 159’s Lean statements express the advertised result?

These assessments compare all the family’s Lean statements with the whole result OpenAI describes in its catalogue. They do not assess this statement alone. Each group uses its own published rules.

Neither group finds that the Lean statements express or imply the whole advertised result. They describe what the statements express in different terms.

MathVet
Finds part of it stated, or a narrower or weaker statement Their term: “partial”.
Only a special case is stated, or a co-equal headline claim has no challenge statement. Their record
Their note: ReciprocalProgressionTheorem : ∀ A : Set ℕ, ¬ Summable (reciprocalTerm A) → ∀ k, HasAP A k (HasAP: a k-term progression a + i·d, d > 0, inside A) is exactly the first claim, stated for all A ⊆ ℕ (0 ∈ A is harmless: it contributes 0 to the sum and the positive-integer case is covered). The paper's quantitative bound |A| ≤ C_k N exp(-c_k (log N)^ε_k) for k-AP-free subsets of {1..N} has no statement (docs: 'outside this statement'); the solution cone is by far the largest in this set (4,774 files, 1,122,983 OAI lines).
Carlos Toledo’s cert-machine
Finds part of it stated, or a narrower or weaker statement Their term: “NARROWER — DECLARED”.
The challenge statements state less than the headline, and OpenAI’s Lean note says so. Their record
Their note: The quantitative Szemerédi bound for k-AP-free subsets of {1,…,N} is not stated.

Read the complete assessments on the family page.

OpenAI’s own note on what this Lean statement covers

“The selected theorem is the reciprocal-sum consequence. The paper's quantitative upper bound for the largest progression-free subset of $\{1,\ldots,N\}$ is outside this statement.”
Quoted from OpenAI’s Lean note for this family

An AI agent drafted an interpretation of that note, classifying this statement as “states a consequence of a paper’s theorem”. No human review of the draft is recorded.

Supporting material

The complete Lean statement file

36 lines, copied from release version fd4aeeb. A name defined in this file links to its line; pointing at the name shows its definition.

Declared in this fileOpenAI’s library at fd4aeebOther linked names open the current documentation for Mathlib, Lean’s mathematical library.

Links in the file: 3 to its own declarations, 4 to Mathlib’s documentation. 4 declared names also link to the same declaration in OpenAI’s library. OpenAI uses Mathlib version d13f23b. The linked documentation describes the current version, so names that have since changed may be missing there.

Definitions and lemmas named by the statement

If Comparator accepts a proof, it proves manuscriptReciprocalProgressionTheorem exactly as written here. To compare that statement with the paper, also read the definitions and lemmas listed below. Atlas found them by following names in the source text and lists each after the names it uses. For x.foo, Atlas includes every foo defined here for a type, even when the statement may use only one of them.

It reaches all 3 other declarations of the file; with the theorem itself they span 9 lines. 1 of them is named in the statement itself.

Names in the namespace OAI.Erdos3 are listed without it.

def HasAPline 7library
def reciprocalTermline 16library
def ReciprocalProgressionTheoremline 24in the statementlibrary
theorem manuscriptReciprocalProgressionTheoremline 31selected theoremlibrary

import Mathlibnamespace OAInamespace Erdos3def HasAP (A : Set ℕ) (k : ℕ) : Prop :=  ∃ a d : ℕ, 0 < d ∧ ∀ i < k, a + i * d ∈ Aend Erdos3open scoped BigOperatorsnamespace Erdos3noncomputable def reciprocalTerm (A : Set ℕ) (n : ℕ) : ℝ := by  classical  exact if n ∈ A then (n : ℝ)⁻¹ else 0end Erdos3namespace Erdos3def ReciprocalProgressionTheorem : Prop :=  ∀ A : Set ℕ, ¬ Summable (reciprocalTerm A) → ∀ k : ℕ, HasAP A kend Erdos3namespace Erdos3theorem manuscriptReciprocalProgressionTheorem : ReciprocalProgressionTheorem := by  sorryend Erdos3end OAI

Every extracted statement of the accompanying papers

The statements the atlas extracted automatically from each accompanying paper’s TeX source, with links to the full sources.

Quasipolynomial Bounds for Arithmetic Progressions

Theorem

For each fixed integer k\ge3 there are constants C_k,c_k,\varepsilon_k>0 such that, for every N\ge2,

r_k(N)\le C_kN\exp\bigl(-c_k(\log N)^{\varepsilon_k}\bigr).

Equivalently, there is A_k\ge1 such that an \alpha-dense subset of [N] contains a nonconstant k-term progression whenever

\log N\ge A_k\bigl(2+\log(1/\alpha)\bigr)^{A_k}, \qquad 0<\alpha\le1.

Section 1, Introduction Read it in the paper TeX source, line 13

Show the TeX

For each fixed integer $k\ge3$ there are constants
$C_k,c_k,\varepsilon_k>0$ such that, for every $N\ge2$,
\begin{equation}
 r_k(N)\le C_kN\exp\bigl(-c_k(\log N)^{\varepsilon_k}\bigr).
\end{equation}
Equivalently, there is $A_k\ge1$ such that an $\alpha$-dense subset
of $[N]$ contains a nonconstant $k$-term progression whenever
\begin{equation}
 \log N\ge A_k\bigl(2+\log(1/\alpha)\bigr)^{A_k},
 \qquad 0<\alpha\le1.
\end{equation}

Corollary (Divergent reciprocal sums)

Every set A\subseteq\mathbb N with \sum_{a\in A}a^{-1}=\infty contains nonconstant arithmetic progressions of every finite length.

Section 1, Introduction Read it in the paper TeX source, line 32

Show the TeX

Every set $A\subseteq\mathbb N$ with
$\sum_{a\in A}a^{-1}=\infty$ contains nonconstant arithmetic
progressions of every finite length.

Configuration and file facts

File: lean/ComparatorChallenges/ErdosReciprocal.lean
OpenAI’s Lean catalogue: OpenAI’s Lean catalogue does not list this as a main result; the release’s Lean notes for family 159 link it.
Solution module: OAI.Combinatorics.Progressions.Main
Allowed axioms: propext, Quot.sound, Classical.choiceThe axioms this configuration allows a proof to use. A replay shows which of them a proof uses.
Imports: Mathlib
Imports from outside Mathlib: None
Local vocabulary: 3 local vocabulary declarations
Lines of Lean, without comments or blank lines: 9
Second proof-verification program (nanoda): Off in OpenAI's configuration.

A shareable summary of this page: the proposed paper match, what the Lean statement expresses, and which reviews are recorded.

Plectis Atlas