Result 229, Probability and statistical mechanics

Exact three- and four-state reconstruction thresholds and four-state tree capacity

Proves the exact reconstruction threshold dλ2>1d\lambda^2\gt 1, with nonreconstruction at equality, for three-state symmetric and four-state ferromagnetic broadcasting on regular trees (d ≥ 2) and observed Poisson trees (mean d > 1 and d > 0, respectively), with Poisson advantage averaged without conditioning on survival. The three-state theorem allows both signs of λ and gives the exact weak-recovery threshold for the symmetric three-community stochastic block model.

Lean formalization Proof

The bigger picture

Why it matters

How much information about an ancestor survives repeated noisy copying? These manuscripts claim exact boundaries for recovering a tree's initial state from distant descendants, and connect one boundary to detecting communities in sparse networks.

What changes?

The manuscripts report reconstruction exactly when d times lambda squared exceeds one, with impossibility at equality. Here d is the branching number or mean offspring count, and lambda measures parent-child signal transmission. This covers symmetric three-state broadcasting with either sign of lambda and ferromagnetic four-state broadcasting, which favors matching states. Regular trees require d at least two; Poisson trees require mean d greater than one and zero, respectively. Poisson trees are observed, and advantage is averaged without conditioning on survival.

What does that help mathematicians do?

For four-state ferromagnetic broadcasting, a further manuscript reports a criterion for any deterministic rooted tree with bounded degree, when lambda lies strictly between zero and one. Reconstruction is equivalent to positive L3 capacity, a measure of the tree's ability to sustain flow, using edge resistances equal to lambda raised to minus twice the edge's depth. This lets researchers assess irregular trees without regularity or growth-rate assumptions, including cases at the exponential critical boundary.

Are there practical applications?

The three-state result, combined with known results, gives a sparse-network community-detection threshold. For three independent, uniformly assigned labels, fixed positive within- and between-community rates a and b, and mean degree (a+2b)/3 > 1, better-than-chance recovery is possible exactly when (a-b)^2 > 3(a+2b). Above it, an O(n log n) algorithm exists; at or below it, recovery is information-theoretically impossible. This is a theoretical guarantee, not evidence of practical performance.

This section was generated by GPT-6 Astra Medium. This explanation is based on the result summary and manuscript abstracts below. This context is separate from OpenAI's source text.

3 manuscripts

The Reconstruction Threshold for the Ferromagnetic Four-State Potts Model

October 5, 2026 22 pages

We establish the exact Kesten–Stigum reconstruction threshold for the ferromagnetic four-state Potts broadcast model on every regular d-ary tree with d ≥ 2 and every Poisson Galton–Watson tree of mean d > 0. Reconstruction occurs exactly when dλ2>1d\lambda^2\gt 1; we prove nonreconstruction at and below the threshold, including equality. In the Poisson model the whole tree is observed and the reconstruction advantage is averaged without conditioning on survival. The proof uses reproducible exact-arithmetic verification of polynomial inequalities.

Cite (BibTeX)
@misc{OAI:The-Reconstruction-Threshold-for-the-Ferromagnetic-Four-State-Potts-Model-October-5-2026,
  author = {{OpenAI}},
  title = {{The Reconstruction Threshold for the Ferromagnetic Four-State Potts Model}},
  howpublished = {OpenAI Math Release preprint
                  \href{https://github.com/openai/math/blob/main/preprints/The-Reconstruction-Threshold-for-the-Ferromagnetic-Four-State-Potts-Model-October-5-2026/four-state-potts.pdf}{OAI:The-Reconstruction-Threshold-for-the-Ferromagnetic-Four-State-Potts-Model-October-5-2026}},
  year = {2026}
}

A Capacity Criterion for Four-State Potts Reconstruction on Trees

October 5, 2026 11 pages

For the ferromagnetic four-state broadcast model with 0<λ<10\lt \lambda\lt 1, we prove that reconstruction on a bounded-degree deterministic rooted tree occurs exactly when its L3 capacity with edge resistances λ−2∣e∣\lambda^{-2|e|} is positive. This gives an exact criterion without regularity or growth-rate assumptions on the tree, including at the exponential critical boundary.

Cite (BibTeX)
@misc{OAI:A-Capacity-Criterion-for-Four-State-Potts-Reconstruction-on-Trees-October-5-2026,
  author = {{OpenAI}},
  title = {{A Capacity Criterion for Four-State Potts Reconstruction on Trees}},
  howpublished = {OpenAI Math Release preprint
                  \href{https://github.com/openai/math/blob/main/preprints/A-Capacity-Criterion-for-Four-State-Potts-Reconstruction-on-Trees-October-5-2026/four-state-capacity.pdf}{OAI:A-Capacity-Criterion-for-Four-State-Potts-Reconstruction-on-Trees-October-5-2026}},
  year = {2026}
}

The exact reconstruction threshold for the three-state symmetric channel

September 25, 2026 60 pages

We determine the exact reconstruction threshold for the symmetric three-state broadcast process on every regular b-ary tree, b ≥ 2, and every observed Poisson Galton–Watson tree of mean d > 1. Reconstruction occurs exactly when dλ2>1d\lambda^2\gt 1, with d = b in the regular model; there is non-reconstruction at equality for either sign of the channel parameter. The Poisson advantage is averaged over trees and spins without conditioning on survival. This resolves the all-degree three-state regular-tree prediction. Combining the Poisson theorem with known tree-to-graph and algorithmic results gives the exact weak-recovery threshold for the symmetric three-community sparse stochastic block model with independent uniform labels, fixed within- and between-community rates a,b>0a,b\gt 0, and mean degree (a+2b)/3>1(a+2b)/3\gt 1: recovery is possible exactly when (a−b)2>3(a+2b)(a-b)^2\gt 3(a+2b). Above this threshold it is achievable in O(nlog⁡n)O(n\log n) time; at or below it, weak recovery is information-theoretically impossible.

Cite (BibTeX)
@misc{OAI:The-exact-reconstruction-threshold-for-the-three-state-symmetric-channel-September-25-2026,
  author = {{OpenAI}},
  title = {{The exact reconstruction threshold for the three-state symmetric channel}},
  howpublished = {OpenAI Math Release preprint
                  \href{https://github.com/openai/math/blob/main/preprints/The-exact-reconstruction-threshold-for-the-three-state-symmetric-channel-September-25-2026/paper.pdf}{OAI:The-exact-reconstruction-threshold-for-the-three-state-symmetric-channel-September-25-2026}},
  year = {2026}
}

Lean formalization

OpenAI's note on what the formalization covers, from lean/docs/229.md.

Exact three- and four-state reconstruction thresholds and four-state tree capacity

The following describes the scope of the Lean formalization related to the following accompanying paper(s):

Scope

The formalization proves the supercritical direction of reconstruction for the symmetric three-state broadcast channel, whose parameter satisfies −1/2≤λ≤1-1/2\le\lambda\le1. On a regular bb-ary tree, reconstruction holds when bλ2>1b\lambda^2>1; on an observed Poisson Galton–Watson tree of mean dd, it holds when dλ2>1d\lambda^2>1. In each case the root-estimation advantage converges to a positive limit, with the Poisson advantage averaged over trees and spins.

The selected statements cover reconstruction above the Kesten–Stigum threshold. Non-reconstruction at or below the threshold and the stochastic-block-model consequences in the paper are outside them.

Comparator links

Result Comparator statement
Supercritical reconstruction on regular and Poisson trees ThreeStateSupercritical.lean
Three-state reconstruction above the Kesten–Stigum threshold ThreeStateTreeClauses.lean

Data from github.com/openai/math at commit adc7f12, committed October 6, 2026 at 21:58 UTC, last checked for changes about 8 hours ago. Titles, subjects, summaries, abstracts and Lean notes are OpenAI's; page counts are read from the PDFs. The map, related results, search, kinds of results and the named-problem index are Emergent Mind's, built with text embeddings and an LLM, and may contain errors.

An Emergent Mind Labs project. Emergent Mind is not affiliated with OpenAI. None of these results has been peer reviewed. Cite the manuscripts themselves, using the BibTeX on each result's page.