All problems

IMO 2026 · Problem 6

Frozen · pending review·Day 2·domain pending·Evidence I0 - Unscoped
Official-solution dependence: None·Official solution viewed: NO
Proof audit: NOT-STARTED·Internal prior-solution exposure: YES

Legacy solution manuscript

A complete solution write-up produced earlier in the project. Preserved here in full; not an official IMO solution and not yet independently audited.

IMO2026_Day2_Problem6.pdfDownload PDF
created 2026-07-16level: L4-candidate (human-checkable write-up)audit: AUDIT-PENDINGsha256:06b2172ce636

Independent HIVE-IMO X manuscript. Not an official IMO solution, and not independently verified under the new pipeline.

Official problem

Statement captured from the official source but not yet passed. Independent English semantic review (Sol, V2) and Human Vietnamese review (V3) are pending; the gate is not yet PASS-OFFICIAL-FREEZE.

Let a1,a2,a3,a_1,a_2,a_3,\ldots be an infinite sequence of positive integers greater than 11. Suppose that for all positive integers nn, the number an+1a_{n+1} is the smallest positive integer greater than ana_n such that gcd(an+1,ai)>1\gcd(a_{n+1},a_i)>1 for every i=1,2,,ni=1,2,\ldots,n. Prove that there exist positive integers TT and LL such that an+T=an+La_{n+T}=a_n+L for every positive integer nn.

(Note that gcd(x,y)\gcd(x,y) denotes the greatest common divisor of positive integers xx and yy.)

official source ↗frozen 2026-07-21sha256:cbe3015874

Vietnamese translation

Faithful literal draft by Fable, pending Human review (proper names kept in original form).

Cho a1,a2,a3,a_1,a_2,a_3,\ldots là một dãy vô hạn các số nguyên dương lớn hơn 11. Giả sử rằng với mọi số nguyên dương nn, số an+1a_{n+1} là số nguyên dương nhỏ nhất lớn hơn ana_n sao cho gcd(an+1,ai)>1\gcd(a_{n+1},a_i)>1 với mọi i=1,2,,ni=1,2,\ldots,n. Chứng minh rằng tồn tại các số nguyên dương TTLL sao cho an+T=an+La_{n+T}=a_n+L với mọi số nguyên dương nn.

(Ở đây gcd(x,y)\gcd(x,y) ký hiệu ước chung lớn nhất của các số nguyên dương xxyy.)

What the problem asks

A short plain-language explanation is written during S1–S3, after independent human and AI reads. S0 freezes the statement only - no interpretation yet.

Human × AI contribution matrix

Attribution per stage. Final approval is always human-led and verified independently.

ContributionHumanSolFableIndependent verifier
Problem interpretation----
Core idea----
Lemma discovery----
Counterexample search----
Proof writing----
Formal checking----
Final approval----
Contribution attribution is recorded stage by stage; no roles are assigned until work begins.

Swipe horizontally to see all columns →

Audit preparation (S0-R)

The manuscript has been reconciled into a verifier-ready dossier: statement binding, section map, and a proof-obligation skeleton. No obligation has been verified.

state: RECONCILEDstatement: MATCHproof obligations: 6 (4 load-bearing)audited: 0/6

Every obligation is marked CLAIMED (asserted by the manuscript), never VERIFIED. Independent audit by an unexposed verifier has not started.

Timeline · S0–S14

  1. S0Official problem freeze
    HUMAN · Pending: Human V3 translation review + contamination attestation.AI · Fable froze the official statement, produced normalized EN + VI draft, source record, contamination ledger and SHA-256 manifest. Sol V2 semantic review pending. No solving.
  2. S1Human blind read
    Not started.
  3. S2AI independent read
    Not started.
  4. S3Contrastive merge
    Not started.
  5. S4Formal decomposition
    Not started.
  6. S5Small-domain reconnaissance
    Not started.
  7. S6Route tournament
    Not started.
  8. S7Lemma forge
    Not started.
  9. S8Candidate proof assembly
    Not started.
  10. S9Adversarial review
    Not started.
  11. S10Independent reconstruction
    Not started.
  12. S11Human proof rewrite
    Not started.
  13. S12Independent / formal verification
    Not started.
  14. S13Official-solution comparison
    Not started.
  15. S14Publication and postmortem
    Not started.

Route map

Up to three main routes survive the tournament at S6.

No routes proposed yet.

Failed attempts

Failed routes and rejected proofs are kept permanently, never deleted.

None recorded yet.

Key lemmas

Lemmas appear here once forged at S7, each with an exact statement and an evidence class.

Proof - compact

Written at S11 in natural olympiad style, scorable 0–7 per step. Not yet available.

Proof - annotated

The annotated proof with justifications and dependencies is not yet available.

Independent verification

Required before the status VERIFIED-INDEPENDENT may be shown.

No independent verification has been performed. Status will not reach VERIFIED-INDEPENDENT until an olympiad expert, an independent model family, or a formal proof (Lean/Isabelle/Coq) confirms the proof.

Official-solution comparison

Official solution viewed after proof freeze: NO
Official solution dependence: NONE

A full comparison (core idea, length, naturalness, differences) runs at S13, only after the proof is frozen.

Artifact downloads

Every file is published with a SHA-256 checksum.

MANIFEST sha256: ce54e4063c4c4a837fc8c7defc6b3ad65850b6c0a5795dfc84d51ff95dffd063

Revision history

  1. 21/07/2026Record initialized
    Ledger created at status FROZEN-PENDING-REVIEW. No mathematical progress recorded yet.

HIVE-IMO X is an independent research project. Not affiliated with or sponsored by the International Mathematical Olympiad. Problem texts belong to their authors. Human Owner Lâm Nguyễn holds academic responsibility; Sol and Fable are AI collaborators, not accountable authors.