Answers are grounded in this paper and cite their sources.
For odd k, let Γ_k = ⟨A, B⟩ ⊂ GL₄ₖ₊₁(ℤ) be the hypergeometric group whose characteristic polynomials are f_k = (x+1)(x⁴ᵏ − x²ᵏ + 1) and g_k = (x²ᵏ − 1)/(x+1) · (x²ᵏ − x^k + 1)(x − 1)². The group Γ_k preserves a rational quadratic form of signature (4k, 1). Its thinness is an instance of Conjecture 2 of Fuchs, Meiri and Sarnak. We prove that Γ_k is thin for every odd k, an infinite family in dimensions n = 4k + 1. The case k = 1 was known by work of Bajpai and Nitsche; the cases k ≥ 3 are not covered by earlier work. The proof shows that the involution σ = −C and the rotation P = −A generate the free product ℤ/2 ∗ ℤ/12k. It does this through a ping-pong set cut out by explicit half-spaces. The construction is canonical and is the same for every k. It uses one base face e = (12k − 1)v + P²ᵏv − P⁴ᵏv, whose boundary contains the geodesic joining the two cusps of the group, together with its 6k − 1 rotates about each cusp under the parabolic elements (σP)±1. A single face witnesses every required containment, and the point v certifies that the ping-pong set is nonempty. The reduction from the certificate to the free product, and from the free product to infinite index, is proved by hand for all k. The certificate is checked in Lean 4 for k ≤ 7 and by exact integer arithmetic for 9 ≤ k ≤ 101. For k ≥ 103 we give a uniform proof. The Gram form factors through (1 + t), and the alternating sums defining the cusp orbits telescope to vectors with at most 14 nonzero entries. This gives a closed form for the cusp pairings and turns every condition into a polynomial inequality of degree at most 3 in (k, s, a) on finitely many polyhedral cells. A computer-assisted case split in exact rational arithmetic settles all 2309 cells. Each cell is shown to contain no lattice points, or is identically zero, or carries an explicit Handelman-type positivity certificate. All 1211 certificates are rechecked independently, and the case split is validated against direct computation at k = 103, 105, 107, 151 and 201. Finally, we prove that no single half-space can serve when k = 3, which explains why the cusp walls are needed. Lean files, code, certificates and logs accompany the paper.
No takes yet. Share an insight, caveat, or question.
William Huanshan Chuang (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: