Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

The Pythagorean Angle Lattice in Quantum Mechanics: Exact Spin Rotations, Gate Synthesis Obstructions and Arithmetic No-Go Theorems for Quantum Computation

Chetansing K. Rajput

Published Oct 10, 2026• 0 citations
Publisher

Abstract

The Pythagorean Angle Lattice produces exact spinor-norm and continuity no-go theorems for single-qubit gate synthesis, free and dense groups of difference gates, and exponent walls at positive distance from the Clifford group. Let θ be the smaller acute angle of a primitive Pythagorean triangle (a, b, c) and z = e^{iθ} = (b + ia)/c. The rational norm-one phases form the Pythagorean angle lattice P ⊂ ℚ(i). We study the exact arithmetic of the associated single-qubit rotations and of the difference angle Δ = π/2 − 2θ, whose spin rotation is U_n(Δ) = (A I − i B n · σ)/(√2 c) with A = a + b, B = b − a. Within the sector of rational rotation matrices, Gaussian rational eigenphases occur exactly for rational Bloch axes, and the obstruction to an exact rational spin lift is the spinor squareclass, which on these phases is the halving class [N(1 + z)]. Every Pythagorean difference gate has spinor norm [2] and factors exactly as a quarter-turn times the rational spin gate defined by the triple. This gives a spinor-norm obstruction to exact synthesis with forced length parity, an obstruction to synthesising the rotation by θ from difference and Clifford gates unless c is a perfect square, and an exact criterion for rational fourth roots; difference gates about orthogonal axes generate a free group, and about non-parallel axes a dense group with approximating words of logarithmic length. For pairs of gates we prove exact commutator identities, including a relation between the commutator scalar and the Born overlap of PAL-internal axis states; nested commutators obey an exact logistic recurrence for their transverse intensity, with a transition at θ = π/12 and a 2-adic filtration of their spectral discriminants. In odd prime-power dimensions, suitable Weil operators admit exponent intertwiners U ↦ U^{−2} at projective distance at least 1/(2√2) and diamond distance at least √(1 − p^{−k}) from the Clifford group, with Weyl spreading at least 1/360. Finally, density of the dyadic classes gives a continuity obstruction: without an a priori height bound, no fixed finite-copy continuous protocol separates them with a uniform margin. All results concern exact arithmetic structures inside ordinary quantum mechanics; no modification of its postulates is proposed.

DOI
10.5281/zenodo.23283559

Related papers

Fermions and Bosons as Geometric Complementarity: Quadratic Coordinates, Rotational Classes, and State Multiplicity

Caroline Johnson

This paper maps the quadratic node sequence r_n = n²r₀ onto the two rotation classes of quantum states. The transition intervals between consecutive nodes are the odd numbers Δr_n = (2n+1)r₀, and they accumulate to the nodes. The interval count 2n+1 is the dimension of the ℓ = n representation, the sum of (2ℓ+1) over ℓ = 0 to n−1 is the orbital degeneracy n² of shell n, and the two spin-1/2 components give the shell capacity 2n². The planar coordinate u = √r e^{iφ/2} has the half-integer rotation character m = 1/2 and changes sign under a 2π rotation, and u² has the integer character m = 1. The half-integer and integer classes are tabulated against the observed particle classes with Particle Data Group spins. The spin-statistics theorem is applied as established physics and is not derived here. The paper does not claim that the geometry causes particle spin, generates the particle classes or produces exchange statistics. The construction rests on Euclid’s Proposition II.5, a² − b² = (a+b)(a−b), which carries the node intervals and the rational rotations through t = tan(ψ/2). Version 2 adds a computational section in which state preparation, gates, measurement and error correction are run along three paths on identical inputs: a complex-matrix path, a scalar path with Taylor-evaluated exponentials, and a Euclid-native path in exact rational arithmetic. For the gate set tested, the Euclid-native amplitudes equal the matrix-path amplitudes up to a positive factor exactly, with zero norm residual and exact outcome probabilities. Real angles enter through continued-fraction convergents with an error of at most 2/q² per gate, and the arithmetic after the convergent is chosen has no rounding. The error-correction tests include the bit-flip, phase-flip and Shor nine-qubit codes, and for the Shor code arbitrary single-qubit errors at real angles with every outcome of nonzero probability followed, with recovery fidelity exactly 1 on every branch. The paths agree to 4×10⁻¹⁵ or better on the original tests and to 8×10⁻¹⁵ or better on the real-angle and arbitrary-error tests. The paper is a map between rotation classes and their algebra. It is not an encoding of fermionic operators onto qubits, and it does not rank the complex-amplitude grid and the discrete algebraic grid, which carry the same relationships on the operations and inputs tested.

•Jul 11, 2025•0 citations

Typical-Case Gate Approximation and Arithmetic Obstructions in Quaternionic Single-Qubit Compilation

Kingsley Yeon, Steven B. Damelin, Greene, Alec

Fault-tolerant quantum computation requires compiling arbitrary one-qubit unitaries into short words over a fixed gate library. For arithmetic libraries such as the $p=5$ Lubotzky--Phillips--Sarnak, or Clifford$+V$, gate set, this problem is governed by quaternionic lattice points on $S^3\cong \SU(2)$. We study the complete norm shells \[ P_k=\{x/5^k\in S^3:x\in\ZZ^4,\ |x|^2=5^{2k}\} \] and the associated projective gate set $T\subset PSU(2)$. The main worst-case quantity is Sarnak's covering exponent $K(T)$, for which the classical range is $4/3\le K(T)\le2$. We show that any positive localized cap-kernel certificate using only the Deligne--LPS square-root spectral estimate reaches only the volume-squared threshold $|V_T(t)|\gg μ(B(\varepsilon))^{-2}$, hence only exponent $2$. Thus any unconditional improvement requires arithmetic cancellation in localized off-diagonal counting. We also record the conditional benchmark that twisted Linnik gives $K(T)=4/3$, matching Harman's obstruction. We prove the shell-to-gate implication $ρ(P_k)\le C5^{-αk}\Rightarrow K(T)\le4/(3α)$, so $α>2/3$ is exactly the threshold for improving the unconditional bound. Finally, exact enumeration of $P_1,P_2,P_3,P_4$ and Haar-random tests show median trace-defect error at the optimal $N^{-2/3}$ scale, while high quantiles remain separated. This gives a sharp distinction between strong typical-case performance and rare arithmetic holes controlling worst-case synthesis.

arXiv:1506.05785•Jun 18, 2015•1 citation

Projective Bloch-Axis Triangles, Quaternion Holonomy, and the Bargmann Invariant

John D. Coady

A pure qubit state is a point of CP¹ ≅ S², determined by an oriented Bloch vector n. Wildberger’s universal hyperbolic geometry (UHG), in its elliptic form, works instead on the real projective plane RP², where n and −n are identified. This is an additional quotient and not the quantum projectivization, because n and −n are orthogonal states. This note asks which parts of the three-state Bargmann invariant are determined by a UHG triangle of Bloch axes, and which require oriented or spinorial data. Main results Quaternionic realization of the Bargmann invariant. Using Wildberger’s projective quaternion description of SO(3), the ordered product K = E₁₂E₂₃E₃₁ of the canonical positive-scalar quaternion representatives of the three shortest edge rotations satisfies K = 8 Φ_{n₁}(Δ₃), where Δ₃ is the Bargmann invariant and Φ_{n₁} embeds the complex numbers in the quaternion plane spanned by 1 and n₁. The oriented base axis n₁ supplies the complex structure. The rotation of K gives only the doubled angle. Quadrea identity. The elliptic UHG quadrea equals one minus the half spread of the rotation represented by the product of the three half-turns, and equals 16 (Im Δ₃)². Eight sign choices. The sign changes nᵢ → ±nᵢ give the same UHG triangle but up to eight distinct physical triples. The paper tabulates which quantities are invariant: quadrances, quadrea and the half-turn rotation are, while the Bargmann value and K take at most eight and four values respectively. A global flip leaves K unchanged and conjugates Δ₃. Rational arithmetic. For rational Bloch vectors, the holonomy rotation has a norm-one rational quaternion representative if and only if |Δ₃|² is a rational square. A worked example shows a rational rotation with no rational unit quaternion representative. The motivation is Freed’s thesis that quantum theory is projective while quantization is linear. The paper uses it to separate three notions: the scalar quotient, the positive-ray double cover, and the spinor-norm obstruction. No claim is made about anomalies, gerbes, or field theory.

•Nov 14, 2026•0 citations

Exact synthesis of single-qubit unitaries over Clifford-cyclotomic gate sets

Simon Forest, David Gosset, Vadym Kliuchnikov +1 more

We generalize an efficient exact synthesis algorithm for single-qubit unitaries over the Clifford+T gate set which was presented by Kliuchnikov, Maslov, and Mosca [Quantum Inf. Comput. 13(7,8), 607–630 (2013)]. Their algorithm takes as input an exactly synthesizable single-qubit unitary—one which can be expressed without error as a product of Clifford and T gates—and outputs a sequence of gates which implements it. The algorithm is optimal in the sense that the length of the sequence, measured by the number of T gates, is smallest possible. In this paper, for each positive even integer n, we consider the “Clifford-cyclotomic” gate set consisting of the Clifford group plus a z-rotation by πn. We present an efficient exact synthesis algorithm which outputs a decomposition using the minimum number of πn z-rotations. For the Clifford+T case n = 4, the group of exactly synthesizable unitaries was shown to be equal to the group of unitaries with entries over the ring Z[eiπn,1/2]. We prove that this characterization holds for a handful of other small values of n but the fraction of positive even integers for which it fails to hold is 100%.

arXiv:1501.04944•Journal of Mathematical Physics•Aug 1, 2015•57 citations

Synthesis and Arithmetic of Single Qutrit Circuits

Amolak Ratan Kalra, Michele Mosca, Dinesh Valluri

In this paper we study single qutrit circuits consisting of words over the Clifford +D cyclotomic gate set, where D=diag(±ξa,±ξb,±ξc) , ξ is a primitive 9 -th root of unity and a,b,c are integers. We characterize classes of qutrit unit vectors z with entries in Z[ξ,1χ] based on the possibility of reducing their smallest denominator exponent (sde) with respect to χ:=1–ξ, by acting an appropriate gate in Clifford +D . We do this by studying the notion of `derivatives mod 3 ' of an arbitrary element of Z[ξ] and using it to study the smallest denominator exponent of HDz where H is the qutrit Hadamard gate and D . In addition, we reduce the problem of finding all unit vectors of a given sde to that of finding integral solutions of a positive definite quadratic form along with some additional constraints. As a consequence we prove that the Clifford +D gates naturally arise as gates with sde 0 and 3 in the group U(3,Z[ξ,1χ]) of 3×3 unitaries with entries in Z[ξ,1χ] . We illustrate the general applicability of these methods to obtain an exact synthesis algorithm for Clifford +R and recover the previously known exact synthesis algorithm of Kliuchnikov, Maslov, Mosca (2012). The framework developed to formulate qutrit gate synthesis for Clifford +D extends to qudits of arbitrary prime power.

arXiv:2311.08696•Quantum•Feb 26, 2025•10 citations

The exact LCU sampling overhead of collective diagonal unitaries: resonances and a continued-fraction dichotomy

Zhihui Wang, Sujit Roy, Manil Maskey +1 more

The minimal sampling overhead of the quadratic collective phase $e^{-iγK^2}$ over one layer of single-qubit rotations is decided by the continued fraction of $γ/π$. Here $K$ is a collective observable with spectrum $\{0,\dots,n\}$, such as the permutation-symmetric Hamming weight. This gate is the cardinality-penalty layer of constrained optimization, the one-axis-twisting gate of spin squeezing and the Kerr phase of a bosonic mode. Instead of compiling it to two-qubit gates, we sample it as an LCU over such layers at overhead $Γ$. Our results give an optimality theory for this overhead. Ancilla-free sampling reproduces the target's outcome probabilities up to the minimal $Γ$, and no smaller factor works for every input; for permutation-symmetric $K$, independent per-qubit angles or one layer of arbitrary single-qubit gates give no further reduction. Rational angles $πp/q$ in lowest terms cost exactly $q$ once $n\ge2q-2$, attained uniquely by a uniform $q$-point sampler; the cost is $Θ(n)$ if and only if $γ/π$ is badly approximable, and bounded in $n$ if and only if it is rational. A second result shows what the symmetry buys: without it, the squaring phase of Fourier arithmetic on $m$ qubits costs at least $2^{0.557m-O(1)}$ over single-qubit gates, beyond the $2^{m/2}$ of every cut bound, by a new certificate-lifting argument. For constructions known only through upper bounds, it gives the optimum: the Fourier-based LCU of Carrera Vazquez, Egger and Woerner is, at fixed $n$, optimal up to a constant at typical angles and beaten by $(n+1)/q$ at rational ones with an $n$-independent sampler; the Kerr decomposition of Upreti, Quesada and Chabaud is the unique optimum over phase shifters at rational parameters without photon-number cutoff; and it identifies the rational couplings where one-axis twisting admits optimal cat decompositions.

quant-pharXiv:2610.03595•Oct 5, 2026•0 citations
The Pythagorean Angle Lattice produces exact spinor-norm and continuity no-go theorems for single-qubit gate synthesis, free and dense groups of difference gates, and exponent walls at positive distance from the Clifford group. Let θ be the smaller acute angle of a primitive Pythagorean triangle (a, b, c) and z = e^{iθ} = (b + ia)/c. The rational norm-one phases form the Pythagorean angle lattice P ⊂ ℚ(i). We study the exact arithmetic of the associated single-qubit rotations and of the difference angle Δ = π/2 − 2θ, whose spin rotation is U_n(Δ) = (A I − i B n · σ)/(√2 c) with A = a + b, B = b − a. Within the sector of rational rotation matrices, Gaussian rational eigenphases occur exactly for rational Bloch axes, and the obstruction to an exact rational spin lift is the spinor squareclass, which on these phases is the halving class [N(1 + z)]. Every Pythagorean difference gate has spinor norm [2] and factors exactly as a quarter-turn times the rational spin gate defined by the triple. This gives a spinor-norm obstruction to exact synthesis with forced length parity, an obstruction to synthesising the rotation by θ from difference and Clifford gates unless c is a perfect square, and an exact criterion for rational fourth roots; difference gates about orthogonal axes generate a free group, and about non-parallel axes a dense group with approximating words of logarithmic length. For pairs of gates we prove exact commutator identities, including a relation between the commutator scalar and the Born overlap of PAL-internal axis states; nested commutators obey an exact logistic recurrence for their transverse intensity, with a transition at θ = π/12 and a 2-adic filtration of their spectral discriminants. In odd prime-power dimensions, suitable Weil operators admit exponent intertwiners U ↦ U^{−2} at projective distance at least 1/(2√2) and diamond distance at least √(1 − p^{−k}) from the Clifford group, with Weyl spreading at least 1/360. Finally, density of the dyadic classes gives a continuity obstruction: without an a priori height bound, no fixed finite-copy continuous protocol separates them with a uniform margin. All results concern exact arithmetic structures inside ordinary quantum mechanics; no modification of its postulates is proposed.
This paper maps the quadratic node sequence r_n = n²r₀ onto the two rotation classes of quantum states. The transition intervals between consecutive nodes are the odd numbers Δr_n = (2n+1)r₀, and they accumulate to the nodes. The interval count 2n+1 is the dimension of the ℓ = n representation, the sum of (2ℓ+1) over ℓ = 0 to n−1 is the orbital degeneracy n² of shell n, and the two spin-1/2 components give the shell capacity 2n². The planar coordinate u = √r e^{iφ/2} has the half-integer rotation character m = 1/2 and changes sign under a 2π rotation, and u² has the integer character m = 1. The half-integer and integer classes are tabulated against the observed particle classes with Particle Data Group spins. The spin-statistics theorem is applied as established physics and is not derived here. The paper does not claim that the geometry causes particle spin, generates the particle classes or produces exchange statistics. The construction rests on Euclid’s Proposition II.5, a² − b² = (a+b)(a−b), which carries the node intervals and the rational rotations through t = tan(ψ/2). Version 2 adds a computational section in which state preparation, gates, measurement and error correction are run along three paths on identical inputs: a complex-matrix path, a scalar path with Taylor-evaluated exponentials, and a Euclid-native path in exact rational arithmetic. For the gate set tested, the Euclid-native amplitudes equal the matrix-path amplitudes up to a positive factor exactly, with zero norm residual and exact outcome probabilities. Real angles enter through continued-fraction convergents with an error of at most 2/q² per gate, and the arithmetic after the convergent is chosen has no rounding. The error-correction tests include the bit-flip, phase-flip and Shor nine-qubit codes, and for the Shor code arbitrary single-qubit errors at real angles with every outcome of nonzero probability followed, with recovery fidelity exactly 1 on every branch. The paths agree to 4×10⁻¹⁵ or better on the original tests and to 8×10⁻¹⁵ or better on the real-angle and arbitrary-error tests. The paper is a map between rotation classes and their algebra. It is not an encoding of fermionic operators onto qubits, and it does not rank the complex-amplitude grid and the discrete algebraic grid, which carry the same relationships on the operations and inputs tested.
Fault-tolerant quantum computation requires compiling arbitrary one-qubit unitaries into short words over a fixed gate library. For arithmetic libraries such as the p=5p=5p=5 Lubotzky--Phillips--Sarnak, or Clifford+V+V+V, gate set, this problem is governed by quaternionic lattice points on $S^3\cong \SU(2)$. We study the complete norm shells \[ P_k=\{x/5^k\in S^3:x\in\ZZ^4,\ |x|^2=5^{2k}\} \] and the associated projective gate set T⊂PSU(2)T\subset PSU(2)T⊂PSU(2). The main worst-case quantity is Sarnak's covering exponent K(T)K(T)K(T), for which the classical range is 4/3≤K(T)≤24/3\le K(T)\le24/3≤K(T)≤2. We show that any positive localized cap-kernel certificate using only the Deligne--LPS square-root spectral estimate reaches only the volume-squared threshold ∣VT(t)∣≫μ(B(ε))−2|V_T(t)|\gg μ(B(\varepsilon))^{-2}∣VT​(t)∣≫μ(B(ε))−2, hence only exponent 222. Thus any unconditional improvement requires arithmetic cancellation in localized off-diagonal counting. We also record the conditional benchmark that twisted Linnik gives K(T)=4/3K(T)=4/3K(T)=4/3, matching Harman's obstruction. We prove the shell-to-gate implication ρ(Pk)≤C5−αk⇒K(T)≤4/(3α)ρ(P_k)\le C5^{-αk}\Rightarrow K(T)\le4/(3α)ρ(Pk​)≤C5−αk⇒K(T)≤4/(3α), so α>2/3α>2/3α>2/3 is exactly the threshold for improving the unconditional bound. Finally, exact enumeration of P1,P2,P3,P4P_1,P_2,P_3,P_4P1​,P2​,P3​,P4​ and Haar-random tests show median trace-defect error at the optimal N−2/3N^{-2/3}N−2/3 scale, while high quantiles remain separated. This gives a sharp distinction between strong typical-case performance and rare arithmetic holes controlling worst-case synthesis.
A pure qubit state is a point of CP¹ ≅ S², determined by an oriented Bloch vector n. Wildberger’s universal hyperbolic geometry (UHG), in its elliptic form, works instead on the real projective plane RP², where n and −n are identified. This is an additional quotient and not the quantum projectivization, because n and −n are orthogonal states. This note asks which parts of the three-state Bargmann invariant are determined by a UHG triangle of Bloch axes, and which require oriented or spinorial data. Main results Quaternionic realization of the Bargmann invariant. Using Wildberger’s projective quaternion description of SO(3), the ordered product K = E₁₂E₂₃E₃₁ of the canonical positive-scalar quaternion representatives of the three shortest edge rotations satisfies K = 8 Φ_{n₁}(Δ₃), where Δ₃ is the Bargmann invariant and Φ_{n₁} embeds the complex numbers in the quaternion plane spanned by 1 and n₁. The oriented base axis n₁ supplies the complex structure. The rotation of K gives only the doubled angle. Quadrea identity. The elliptic UHG quadrea equals one minus the half spread of the rotation represented by the product of the three half-turns, and equals 16 (Im Δ₃)². Eight sign choices. The sign changes nᵢ → ±nᵢ give the same UHG triangle but up to eight distinct physical triples. The paper tabulates which quantities are invariant: quadrances, quadrea and the half-turn rotation are, while the Bargmann value and K take at most eight and four values respectively. A global flip leaves K unchanged and conjugates Δ₃. Rational arithmetic. For rational Bloch vectors, the holonomy rotation has a norm-one rational quaternion representative if and only if |Δ₃|² is a rational square. A worked example shows a rational rotation with no rational unit quaternion representative. The motivation is Freed’s thesis that quantum theory is projective while quantization is linear. The paper uses it to separate three notions: the scalar quotient, the positive-ray double cover, and the spinor-norm obstruction. No claim is made about anomalies, gerbes, or field theory.
We generalize an efficient exact synthesis algorithm for single-qubit unitaries over the Clifford+T gate set which was presented by Kliuchnikov, Maslov, and Mosca [Quantum Inf. Comput. 13(7,8), 607–630 (2013)]. Their algorithm takes as input an exactly synthesizable single-qubit unitary—one which can be expressed without error as a product of Clifford and T gates—and outputs a sequence of gates which implements it. The algorithm is optimal in the sense that the length of the sequence, measured by the number of T gates, is smallest possible. In this paper, for each positive even integer n, we consider the “Clifford-cyclotomic” gate set consisting of the Clifford group plus a z-rotation by πn. We present an efficient exact synthesis algorithm which outputs a decomposition using the minimum number of πn z-rotations. For the Clifford+T case n = 4, the group of exactly synthesizable unitaries was shown to be equal to the group of unitaries with entries over the ring Z[eiπn,1/2]. We prove that this characterization holds for a handful of other small values of n but the fraction of positive even integers for which it fails to hold is 100%.
In this paper we study single qutrit circuits consisting of words over the Clifford +D cyclotomic gate set, where D=diag(±ξa,±ξb,±ξc) , ξ is a primitive 9 -th root of unity and a,b,c are integers. We characterize classes of qutrit unit vectors z with entries in Z[ξ,1χ] based on the possibility of reducing their smallest denominator exponent (sde) with respect to χ:=1–ξ, by acting an appropriate gate in Clifford +D . We do this by studying the notion of `derivatives mod 3 ' of an arbitrary element of Z[ξ] and using it to study the smallest denominator exponent of HDz where H is the qutrit Hadamard gate and D . In addition, we reduce the problem of finding all unit vectors of a given sde to that of finding integral solutions of a positive definite quadratic form along with some additional constraints. As a consequence we prove that the Clifford +D gates naturally arise as gates with sde 0 and 3 in the group U(3,Z[ξ,1χ]) of 3×3 unitaries with entries in Z[ξ,1χ] . We illustrate the general applicability of these methods to obtain an exact synthesis algorithm for Clifford +R and recover the previously known exact synthesis algorithm of Kliuchnikov, Maslov, Mosca (2012). The framework developed to formulate qutrit gate synthesis for Clifford +D extends to qudits of arbitrary prime power.
The minimal sampling overhead of the quadratic collective phase e−iγK2e^{-iγK^2}e−iγK2 over one layer of single-qubit rotations is decided by the continued fraction of γ/πγ/πγ/π. Here KKK is a collective observable with spectrum {0,…,n}\{0,\dots,n\}{0,…,n}, such as the permutation-symmetric Hamming weight. This gate is the cardinality-penalty layer of constrained optimization, the one-axis-twisting gate of spin squeezing and the Kerr phase of a bosonic mode. Instead of compiling it to two-qubit gates, we sample it as an LCU over such layers at overhead ΓΓΓ. Our results give an optimality theory for this overhead. Ancilla-free sampling reproduces the target's outcome probabilities up to the minimal ΓΓΓ, and no smaller factor works for every input; for permutation-symmetric KKK, independent per-qubit angles or one layer of arbitrary single-qubit gates give no further reduction. Rational angles πp/qπp/qπp/q in lowest terms cost exactly qqq once n≥2q−2n\ge2q-2n≥2q−2, attained uniquely by a uniform qqq-point sampler; the cost is Θ(n)Θ(n)Θ(n) if and only if γ/πγ/πγ/π is badly approximable, and bounded in nnn if and only if it is rational. A second result shows what the symmetry buys: without it, the squaring phase of Fourier arithmetic on mmm qubits costs at least 20.557m−O(1)2^{0.557m-O(1)}20.557m−O(1) over single-qubit gates, beyond the 2m/22^{m/2}2m/2 of every cut bound, by a new certificate-lifting argument. For constructions known only through upper bounds, it gives the optimum: the Fourier-based LCU of Carrera Vazquez, Egger and Woerner is, at fixed nnn, optimal up to a constant at typical angles and beaten by (n+1)/q(n+1)/q(n+1)/q at rational ones with an nnn-independent sampler; the Kerr decomposition of Upreti, Quesada and Chabaud is the unique optimum over phase shifters at rational parameters without photon-number cutoff; and it identifies the rational couplings where one-axis twisting admits optimal cat decompositions.