← All problems

Saxl's staircase tensor-square conjecture

2 years old · documented by

For k≥1k\geq 1, let

δk=(k,k−1,…,1)\delta_k=(k,k-1,\ldots,1)

be the staircase partition, which has size (k+12)\binom{k+1}{2}. For partitions of this size, g(δk,δk,μ)g(\delta_k,\delta_k,\mu) is the multiplicity of the irreducible S(k+12)S_{\binom{k+1}{2}}-module indexed by μ\mu in the tensor square of the module indexed by δk\delta_k.

Saxl's conjecture.

g(δk,δk,μ)>0g(\delta_k,\delta_k,\mu)>0

for all kk and all μ⊢(k+12)\mu\vdash\binom{k+1}{2}.

This is a concrete proposed choice of the symmetric partition in the tensor square conjecture. Its significance is that the staircase partition is specified uniformly rather than existentially.

References
  1. Greta Panova, “Complexity and asymptotics of structure constants”, arXiv:2305.02553 (2023).

Progress summary

Refreshed
Open · progress reported

The conjecture remains open: earlier partial results stand, while a claimed proof was withdrawn and a new conditional argument is unverified.

Saxl conjectured that the tensor square of the staircase representation contains every irreducible representation of the corresponding symmetric group. The conjecture is known computationally in small cases and through several broad families, but not in full generality.

Known results

  • Bessenrodt and Bowman (2023): computational evidence and several constituent families, without a full proof.
  • Pak, Panova, and Vallejo: sufficient conditions proving positivity for several classes of partitions.
  • A separate positivity paper proves additional families of constituents, but does not establish the conjecture universally.

December 2025 withdrawn proof

Soong Kyum Lee claimed an unconditional proof, but the arXiv record says the preprint was withdrawn after reviewers identified mathematical gaps; the complete-proof claim is therefore unverified. A November 7, 2026 community submission gives a conditional lower bound implying positivity, explicitly assuming source theorems and not providing an unconditional resolution.

Community submission (unverified)

Posted November 7, 2026: a submitted argument claims that, conditional on stipulated source theorems, every nontrivial constituent has multiplicity at least k−1k-1, while the trivial and sign constituents have multiplicity one. It explicitly states that this is not an unconditional proof.

Current status (as of November 2026): Saxl's conjecture remains open; published partial results and an unverified conditional community submission do not resolve it, and the December 2025 claimed proof was withdrawn.

Sources

Solutions 1

Conditional quantitative partial progress on Saxl: assuming specified cyclic-polytabloid source theorems, the nontrivial/nonsign multiplicities are ≥m−1 for every m≥2.See full solutionHide full solution

Conditional scope. For the staircase ρ_m=(m,m−1,…,1), n=m(m+1)/2 and m≥2, the paper derives g(ρ_m,ρ_m,λ)≥m−1 for every λ other than (n) and (1^n), under its specified pinned cyclic-polytabloid source theorems. Equality occurs at λ=(n−1,1). The trivial and sign constituents have multiplicity one by self-duality and staircase self-conjugacy; m=1 is immediate. Thus the conditional estimate implies positivity for all constituents. The new deduction evaluates an inherited recursion; it is not an independent proof of the stipulated source theorems or an unconditional resolution of Saxl’s conjecture.

Paper. The sharp minimum multiplicity in a staircase tensor square. Supporting proof and verification archive.

Source assumptions and verification limits. The identified OpenAI source inputs at immutable revision fd4aeeb2ee4fc729c18d98444fed42fd0529eeeb are stipulated true, as explicitly stated in the preprint. This entry preserves that assumption and is labeled partial progress, not a complete unconditional solution. Lean checks scoped supporting results only, not the full theorem. AI-assisted preprint; internal review is not external peer review, MathDB endorsement or certification of worldwide priority.