--- title: Quantum Pyramids Conjecture Overview url: https://www.emergentmind.com/topics/quantum-pyramids-conjecture type: topic --- # Quantum Pyramids Conjecture Overview Searching arXiv for the cited works and closely related context papers. “Quantum Pyramids Conjecture” is not a standard term with a single settled meaning in the arXiv literature. In current usage it refers to several distinct “pyramid” programs in quantum theory: an informal moniker for the Quantum PCP Conjecture in Hamiltonian complexity; a conjecture, now resolved, about the accessible information of equiangular equiprobable pure-state ensembles called quantum pyramids; a conjectural organizing principle for massless observables in Schrödinger CFTs via reducible but indecomposable “pyramid” modules built from “alien operators”; and, in magnetic resonance, conjectures motivated by the “quantum Pascal pyramid” for multispin intensity patterns [1309.7495] [2606.14698] [2606.29539] [2404.03560]. The phrase therefore denotes a family of domain-specific conjectures rather than a unique theorem-sized statement. ## 1. Terminological scope and principal usages Because the phrase is nonstandard, its meaning is determined by context. In Hamiltonian complexity, it is used informally for the Quantum PCP Conjecture. In quantum information theory, it names a concrete conjecture about optimal POVMs for symmetric state ensembles. In Schrödinger CFT, it is a natural shorthand for the conjectural universality of pyramid representations in the massless sector. In spin dynamics, it labels conjectures suggested by the Jacobi-polynomial and Hermite–Gaussian structure of the “quantum Pascal pyramid” [1309.7495] [2606.14698] [2606.29539] [2404.03560]. | Usage | Core object | Status | |---|---|---| | Hamiltonian complexity | Constant-gap QMA-hardness / quantum games PCP | Open | | Symmetric state discrimination | Information-optimal POVM for equiangular ensembles | Resolved | | Schrödinger CFT | Massless observables as pyramid modules | Partly proven, partly conjectural | | Magnetic resonance | Jacobi/Hermite–Gaussian multispin structure | Conjectural | A common misconception is that all appearances of “quantum pyramid” refer to the same problem. The literature instead supports a disambiguated reading: the shared word “pyramid” names distinct mathematical structures—PCP hardness landscapes, symmetric pure-state ensembles, staggered modules in non-relativistic conformal representation theory, and combinatorial arrays of multispin coefficients. ## 2. The Hamiltonian-complexity interpretation: Quantum PCP and quantum games PCP In the Hamiltonian-complexity literature, “Quantum Pyramids Conjecture” is best understood as an informal label for the Quantum PCP Conjecture. The standard Hamiltonian formulation asks for constant-gap hardness of the Local Hamiltonian problem, QMA’s complete problem. Writing $$ H=\sum_{i=1}^m H_i, $$ with each \(H_i\) acting nontrivially on at most \(k\) qubits and satisfying \(\|H_i\|\le 1\), the promise problem is to decide whether \(\lambda_{\min}(H)\le a\) or \(\lambda_{\min}(H)\ge b\), where \(b-a\ge 1/\mathrm{poly}(n)\). The qPCP conjecture asks whether there exists \(\Delta>0\) such that deciding whether \(\lambda_{\min}(H)\le A\) or \(\lambda_{\min}(H)\ge A+\Delta\), under suitable normalization, is QMA-hard; equivalently, whether QMA admits constant-query proof verification with a constant completeness–soundness gap [1309.7495] [2403.13084]. The 2024 status note emphasizes that the classical equivalence between constraint-satisfaction PCPs and games PCPs does not carry over naively to the quantum setting. Ji, Natarajan, Vidick, Wright, and Yuen proved \( \mathrm{MIP}^*=\mathrm{RE} \), and Natarajan–Zhang showed that even polylogarithmic message lengths retain RE power: $$ \mathrm{MIP}^*[q=O(1),a=\mathrm{polylog}(n)]=\mathrm{RE}. $$ Since RE includes undecidable problems such as the halting problem, any meaningful “quantum games PCP” must restrict the provers to computationally efficient strategies, such as QPT provers, and possibly impose structural constraints on allowed states or measurements [2001.04383] [2302.04322] [2403.13084]. Within that restricted setting, a concrete positive result is available at the AM level. The 2024 note constructs a succinct two-prover one-round MIP\(^*\) protocol for a canonical AM-complete problem with \(O(\mathrm{polylog}(n))\)-bit questions and answers and a constant completeness–soundness gap. The protocol combines introspection, using self-testing and low-degree-test techniques to force the provers to “sample their own questions,” with PCPP-style answer reduction, in which Bob supplies a PCP of Proximity for the predicate \(V(x,r,w(r))\). Completeness can be made \(1\) (or \(1-\epsilon\) due to test slack), and soundness at most \(1/3\) (or \(1/3+\epsilon'\)) [2403.13084]. The same note also identifies a flaw in Natarajan–Vidick’s proposed “games PCP for QMA.” For Pauli-sum Hamiltonians $$ H=\sum_{P\in S}\beta_P P, $$ the relevant normalization for measurement-based energy tests is the Pauli \(1\)-norm, $$ \|H\|_{P,1}:=\min \sum_{P\in S} |\beta_P|, $$ not the operator norm. The acceptance probability satisfies $$ \Pr[\mathrm{accept}]=\frac12+\frac{\langle H\rangle_\psi}{2\|H\|_{P,1}}. $$ Under the proposed amplification $$ H' := 2\Bigl(\frac{I+H}{2}\Bigr)^{\otimes k}-I, $$ one has $$ \|H'\|_{P,1}\le 1+2\Bigl(\frac{1+\|H\|_{P,1}}{2}\Bigr)^k, $$ so exponential growth of the Pauli \(1\)-norm can erase the apparent constant gap. The claim that this amplification yields a QMA-level games PCP is therefore invalid, even though the underlying gap-preserving protocol for Pauli-sum Hamiltonians remains useful [1710.03062] [2403.13084]. This isolates the main open direction. A quantum games PCP for QMA would require gap amplification that preserves or controls the Pauli spectrum, especially \(\|H\|_{P,1}\), together with stronger self-tests and efficiently implementable honest strategies. The note highlights as central open problems the existence of QMA-complete Hamiltonian families with bounded Pauli \(1\)-norm, negative results for highly constrained Pauli spectra, bounds on \(\|f(H)\|_{P,1}\) for polynomial functional amplification, and a two-prover succinct MIP\(^*\) protocol for QMA with efficient provers [2403.13084]. ## 3. Pointer QPCP, Set Local Hamiltonian, and restricted game equivalences A related but distinct line of work replaces the standard qPCP statement by the Pointer QPCP conjecture. In this model, a proof has a classical part \(y=(y_1,\dots,y_m)\), where each \(y_i\in [l(n)]\), and a quantum part \(|\Psi\rangle\) on \(p(n)\) qubits. The verifier chooses uniformly random \(i\in [m]\), reads \(y_i\), and on the basis of \(x\), its internal randomness, and \(y_i\), selects \(q=O(1)\) qubits of \(|\Psi\rangle\) on which it performs a POVM. The conjecture states $$ \mathrm{QMA}=\mathrm{PQPCP}(q,\alpha,\beta) $$ with \(q=O(1)\) and \(\alpha-\beta=\Omega(1)\) [1603.00903]. The corresponding constraint problem is Set Local Hamiltonian. The input consists of \(m\) sets of \(k\)-local terms, $$ \mathcal{H}_i=\{H_{i,1},\dots,H_{i,l(n)}\}, $$ with \(0\le H_{i,j}\le I\). For a selector \(f:[m]\to [l(n)]\), define $$ H_f=\sum_{i=1}^m H_{i,f(i)}. $$ The promise is to distinguish the case in which there exists \(f\) and \(|\phi\rangle\) with \(\langle \phi|H_f|\phi\rangle\le a\,m\) from the case in which for all \(f\) and \(|\phi\rangle\), \(\langle \phi|H_f|\phi\rangle\ge b\,m\). With inverse-polynomial gap this problem is QMA-complete; the conjectural constant-gap version is equivalent to Pointer QPCP [1603.00903]. The game-theoretic counterpart is a restricted class of multi-prover games called CRESP games. These use one classical prover and \(R=\lceil \log(n+1)\rceil\) quantum provers. The quantum provers share an encoding of an arbitrary \(n\)-qubit state under a fixed isometry, and their strategies are restricted to “swap” answers: they may only select and return designated shares of the pre-shared encoding. The verifier performs a codespace test \(T1\), a fair-coin step \(T2\), and then a decoded local-term check \(T3\). The value \(\omega(G)\) is the supremum of the verifier’s acceptance probability over legal strategies [1603.00903]. The main theorem of this framework is a three-way equivalence: the Pointer QPCP conjecture, constant-gap QMA-completeness of Set Local Hamiltonian, and constant-gap hardness of approximating CRESP game value are either all true or all false. This was presented as the first equivalence between a quantum PCP-type statement and inapproximability of a polynomial-size quantum multi-prover game. At the same time, the framework is explicitly weaker than standard QPCP, because the classical pointer can encode per-query strategy information; whether Pointer QPCP implies the original qPCP remains open [1603.00903]. The significance of this formulation is methodological. It offers a game-based PCP analogue without invoking the unrestricted entangled-prover power that leads to RE. A plausible implication is that restricted encodings, rather than unconstrained nonlocality, are the natural setting in which a quantum games PCP may be compared to Local Hamiltonian hardness. ## 4. Quantum pyramids in state discrimination In quantum information theory, a “quantum pyramid” is a specific equiprobable ensemble of \(d\) pure states whose edges are equiangular. Fixing an orthonormal basis \(\{|e_1\rangle,\dots,|e_d\rangle\}\) and the axis vector \(|e_0\rangle \equiv d^{-1/2}\sum_j |e_j\rangle\), the states can be written as $$ |\psi_j\rangle=\sqrt{d r_1}\,|e_j\rangle+(\sqrt{r_0}-\sqrt{r_1})\,|e_0\rangle, $$ with \(r_0,r_1\ge 0\) and \(r_0+(d-1)r_1=1\). Equivalently, $$ |\psi_j\rangle=a\,|e_j\rangle+b\sum_{i\neq j}|e_i\rangle,\qquad a^2+(d-1)b^2=1, $$ and for \(j\neq k\), $$ \langle \psi_j|\psi_k\rangle=\xi=r_0-r_1=2ab+(d-2)b^2. $$ The regimes are classified as acute, obtuse, and flat according to \(\xi\) [2606.14698]. Englert and Řeháček conjectured that the globally information-optimal POVM for this ensemble has a regime-dependent form. The 2026 paper proves the remaining entropy inequalities and thereby resolves the conjecture completely. In the acute case, the optimal POVM is a rank-1 projective measurement onto a deformed computational basis \(\{|\tilde e_k\rangle\}\) with parameter \(t\in [0,1]\). In the obtuse case, the optimal POVM is a two-block measurement mixing the same deformed basis with pairwise difference projectors $$ |ij\rangle=\frac{|e_i\rangle-|e_j\rangle}{\sqrt{2}}, $$ with parameter \(t\ge 1\). In the flat case, the same two-block family is optimal, with a dimension-dependent transition: \(t_F^*(d)=\infty\) for \(3\le d\le 6\) and \(t_F^*(d)=1\) for \(d\ge 7\) [2606.14698]. The accessible information of an equiprobable ensemble \(\mathcal{E}=\{\frac1d,|\psi_j\rangle\langle \psi_j|\}\) under a POVM \(\mathcal{M}=\{M_k\}\) is $$ I(X:Y)=\sum_{j,k}\frac{1}{d}\,p(k|j)\,\log_2\frac{p(k|j)}{\sum_{j'}\frac{1}{d}p(k|j')}, \qquad p(k|j)=\langle \psi_j|M_k|\psi_j\rangle. $$ The final theorem gives a piecewise formula for the optimum \(A(\mathcal{E})\). On the SRM branch, $$ A(\mathcal{E})=\log_2 d + p\log_2 p +(d-1)q\log_2 q. $$ In the flat case, $$ A(\mathcal{E})= \begin{cases} \log_2(d/2), & 3\le d\le 6,\\[2pt] \frac{d-2}{d}\log_2(d-1), & d\ge 7. \end{cases} $$ For \(d=3\), the flat limit gives \(A(\mathcal{E})=\log_2(3/2)\approx 0.585\) bits [2606.14698]. The proof has two distinct technical cores. For obtuse pyramids, the key step is to show that local minimizers of the relevant entropy inequality cannot have exactly three distinct coordinate values. This reduces to a reciprocal inequality involving the principal and lower real branches of the Lambert \(W\) function. For flat pyramids, the essential input is a sharp \(\ell^p\) inequality for zero-sum vectors, proved for all \(d\ge 2\) by the equal variables method. The result confirms the globally information-optimal measurement for all acute, obtuse, and flat quantum pyramids, so in this usage the “Quantum Pyramids Conjecture” is no longer conjectural but a theorem [2606.14698]. ## 5. Pyramid representations and alien operators in Schrödinger CFTs In Schrödinger CFT, the phrase refers to a conjectural organizing principle for the massless sector of observables. The symmetry group in \((d+1)\) dimensions is $$ S_d := (SL(2,\mathbb{R})\times SO(d))\ltimes H_d, $$ generated by \(P_0,D,C_0,M_{ij},P_i,K_i\), and a central mass element \(M\). In the massless sector \(M=0\), the commutator \([K_i,P_j]\propto M\) vanishes, so \(K_i\) and \(P_j\) commute. The physical Hamiltonian in the harmonic-trap picture is $$ H_{\mathrm{HT}}:=P_0+\omega^2 C_0, $$ and the doubled state-operator correspondence identifies local operators with states in \(\mathcal{H}\otimes \mathcal{H}^*\) [2606.29539]. The paper’s central structural claim is that normal-ordered zero-mass observables are organized into reducible but indecomposable “pyramid representations.” These are built from a lowest-weight massless primary \(v_0\) together with an infinite ladder of “alien” states \(v_k\), defined intrinsically by a direct limit of short exact sequences $$ 0\to \mathcal{S}_{k-1}\to \mathcal{S}_k\to \mathcal{V}_{\Delta+k,0}\to 0,\qquad k\ge 1, $$ with lifts satisfying $$ D v_k=(\Delta+k)v_k,\qquad Mv_k=0,\qquad C_0 v_k=0,\qquad K_1 v_k=v_{k-1}. $$ The module has filtration $$ \mathcal{V}_{\Delta,0}=\mathcal{S}_0\subset \mathcal{S}_1\subset \cdots \subset \mathcal{S}_\infty=\mathcal{S}_\Delta, $$ and level-\(k\) dimension $$ \dim (\mathcal{S}_\Delta)_k=\frac{(k+2)^2}{4}. $$ Alien operators are states not killed by \(K_i\) but also not descendants under the diagonal raising; they populate the “alien side” of the pyramid [2606.29539]. A key construction comes from tensor products of opposite-mass modules: $$ \mathcal{L}_{\Delta_1,m}\otimes \mathcal{L}_{\Delta_2,-m} = \bigoplus_\ell p_\ell\,\mathcal{S}_{\Delta_1+\Delta_2+2\ell}. $$ In the explicit \((1+1)\)-dimensional construction, anti-diagonal time derivatives \(\widetilde P_0\) raise the tip dimension by \(2\) per step, while anti-diagonal spatial derivatives \(\widetilde P_1\) generate alien operators. For generic \(\Delta_i\) and \(m\), one finds \(p_\ell=1\). This underlies the claim that the massless \(M=0\) sector of normal-ordered observables decomposes into pyramid modules \(\mathcal{S}_\Delta\) [2606.29539]. The analogy proposed in the paper is with double-twist towers in Lorentzian CFT. The \(s\)-channel pyramid tips sit at $$ \Delta=\Delta_1+\Delta_2+2\ell, $$ and their crossed-channel behavior is governed by \(SL(2,\mathbb{R})\) blocks $$ G_\Delta(s):=s^{\Delta/2}\,{}_2F_1(\Delta/2,\Delta/2;\Delta;-s). $$ At large \(s\), \(G_\Delta(s)\sim A_\Delta \log s + B_\Delta + \cdots\), so genuine massless exchange induces logarithms whose resummation shifts dimensions \(\Delta\to \Delta+\gamma\). In the absence of genuine massless operators, dagger-undagger OPEs are regular, four-point functions factorize, and the normal-ordered composites are non-renormalized [2606.29539]. The pyramid formalism also generalizes exceptional conservation laws. For a conserved current, $$ P_0 J^0 + P_i J^i = 0, $$ with \(\Delta_{J^0}=d\) and \(\Delta_{J^i}=d+1\), one obtains $$ K_i J^i = -\Delta_{J^0}J^0. $$ Thus \(J^0\) lies one alien rung below \(J^i\) in the same pyramid. Similar descent relations hold for higher-rank conserved tensors, and the thermofield-double state decomposes into pyramid “thermal blocks,” which the paper proposes as a natural basis for non-relativistic hydrodynamics and EFT computations [2606.29539]. The status is mixed. The algebraic constructions, tensor-product decompositions, SES definition, filtration, dimension formula, and \(K_i\)-surjectivity are established. The universality claim—namely, that the entire \(M=0\) sector of observables in generic Schrödinger CFTs decomposes into pyramid modules with universal cross-channel renormalization—is conjectural, though supported by explicit constructions, Ward identities, OPE consistency, and non-renormalization arguments [2606.29539]. ## 6. The quantum Pascal pyramid and conjectures in spin dynamics In magnetic resonance, the “quantum Pascal pyramid” is a three-index array relating multispin longitudinal operators and population operators in \(I_N S\) systems with \(N\) identical spin-\(1/2\) nuclei \(I\) symmetrically coupled to a single probe spin \(S\). For each product rank \(q\), $$ \hat Z_N^q \equiv 2^q \sum_{\{i_1<\dots