A 2024 construction by Z. Wang (Phys. Rev. B 111, 104315 (2025)) built a family of exactly solvable one-dimensional unitary quantum circuits from finite-dimensional bialgebras and Hopf algebras, proving that any time-evolved local operator is an exact matrix product operator (MPO) of finite bond dimension and leaving two precisely stated conjectures open. The first (the bialgebra converse) asserts that every finite-dimensional solution of the defining tensor identities of the framework (the pentagon-type system of Eq. (1) of that paper) arises from a finite-dimensional bialgebra via its Eq. (2), i.e., that bialgebras generate all solvable circuits of this type. The second asserts that for every finite-dimensional C*-weak Hopf algebra the canonical element satisfies cη+ν=cν\mathbf{c}^{\eta+\nu} = \mathbf{c}^{\nu} for some positive integers η,ν\eta, \nu (the finite-exponent property), which would give linear-in-system-size revival times for the non-dual-unitary circuits built from weak Hopf algebras; it is verified only numerically for the Fibonacci/Lee-Yang example (with η=5\eta = 5, ν=2\nu = 2). Both conjectures remain open. Resolving them would complete the algebraic characterization of this framework: exhaustiveness of the bialgebra classification of solvable gates, and finite-revival structure for weak-Hopf circuits.

Background

Exactly solvable models are the benchmark laboratories of quantum many-body dynamics: they allow rigorous study of thermalization, entanglement growth, scrambling, and revival phenomena that are generically inaccessible. A prominent modern setting is that of discrete-time unitary (Floquet brickwork) circuits, in which two-site gates are applied in a checkerboard pattern in space-time; notable exactly solvable families include dual-unitary circuits (built from two-site gates that are unitary both along the space direction and along the time direction of the space-time lattice) and the quantum Rule 54 cellular automaton (a Floquet circuit based on the classical elementary cellular automaton rule 54). The key computational tool is the tensor-network representation: a matrix product state (MPS) is a one-dimensional quantum state built from repeated local tensors with an auxiliary 'bond' index whose range (the bond dimension) controls the cost of exact computations, and a matrix product operator (MPO) is the operator analogue. If every time-evolved local operator admits an exact MPO with finite bond dimension, the entire dynamics is exactly computable for any initial MPS.

Wang's construction supplies such a mechanism from algebraic data. A bialgebra is an associative algebra A equipped with a compatible coproduct (comultiplication) Delta: A -> A tensor A and counit epsilon; a Hopf algebra additionally has an antipode, and a C*-Hopf algebra is a finite-dimensional one carrying a compatible involution, which in this setting makes it a direct sum of matrix algebras. A representation of A is a matrix realization rho: A -> Mat_d(C); a corepresentation is a matrix of algebra elements {v_ij in A} satisfying Delta(v_ij) = sum_k v_ik tensor v_kj. Wang proves (his Theorem 1) that any triple {A, rho, v} generates a solution of a closed system of diagrammatic tensor identities (Eq. (1) of that paper) relating three four-index tensors — the gate tensor W (with indices i, j and a, b), the tensor rho (with indices a, b together with two basis indices), and the tensor v (with indices i, j together with two basis indices) — together with boundary covectors: W^{ab}_{ij} = rho_ab(v_ij), the rho-tensor is (delta_y tensor rho_ab) applied to Delta(x), the v-tensor is delta_y(x v_ij), and the boundaries are the unit and counit, where {x} is a basis of A and {delta_x} the dual basis of A*. The first identity is a pentagon-type exchange relation (a linear form of the pentagon equation familiar from conformal field theory and Hopf-algebra duality); the remaining ones are unit/counit normalizations. Given any solution of this system, the two-dimensional tensor network built from W (the space-time 'pyramid' of the circuit) admits an exact MPO representation for time-evolved local operators, with finite bond dimension for all times. With C*-Hopf data the gate W is unitary, yielding exactly solvable unitary circuits; in the finite-dimensional C*-Hopf case the Larson-Radford theorem (the antipode of a finite-dimensional C*-Hopf algebra is involutive) additionally forces every such gate to be dual-unitary. Weak Hopf algebras, in which Delta(1) is not 1 and the counit is multiplicative only on two distinguished subalgebras, escape this constraint: a C*-weak Hopf algebra with Fibonacci anyon fusion rules (the Böhm-Szlachányi Lee-Yang quantum group) produces a non-dual-unitary gate closely related to a Floquet version of the PXP model (a prototypical constrained Rydberg-blockade spin chain), solvable on a constrained subspace.

For finite chains with periodic boundary conditions (PBC), the revival time (the period of the evolution operator) is controlled by the exponent of the underlying algebra: by a theorem of Etingof and Gelaki, for any finite-dimensional semisimple cosemisimple Hopf algebra (in particular any C*-Hopf algebra) the canonical element c = sum_x x tensor delta_x in A tensor A* has finite order, c^eta = 1 tensor epsilon, so t_rev = eta L for L-site chains. For finite-dimensional C*-weak Hopf algebras the canonical element need not be invertible, and Wang states as a conjecture the eventual-periodicity form c^(eta+nu) = c^nu, verified numerically only for the Fibonacci example.

Two bodies of adjacent mathematics make the conjectures plausible but do not touch them. Set-theoretic solutions of the pentagon equation have been classified (bijective solutions decompose as a semigroup of left zeros times a group, and finite solutions correspond to Hopf algebras with a positive basis), but this covers only monomial solutions, far smaller than the general linear tensor system of Eq. (1). The exponent theory of Hopf algebras is active (Etingof-Gelaki and later work on non-cosemisimple cases), but no notion of exponent of a weak Hopf algebra has been developed in the mathematics literature; the tensor-category background for weak Hopf algebras is systematized in the monograph of Etingof, Gelaki, Nikshych and Ostrik. Related MPO-algebra technology for weak Hopf algebras and projected entangled-pair states was developed by Molnár, Ruiz-de-Alarcón, Garre-Rubio, Schuch, Cirac and Pérez-García.

Problem Statement

Decide the two conjectures left open by the Hopf-algebraic construction of exactly solvable unitary circuits, in the original generality of the source (Z. Wang, Phys. Rev. B 111, 104315 (2025)); both components are required for completion, and each is to be settled independently, by rigorous proof or by explicit finite counterexample.

Component (i) — the bialgebra converse conjecture. Consider the tensor identities of Eq. (1) of the source: finite-dimensional tensors — the four-index gate tensor WW (with indices i,ji,j and a,ba,b), the four-index tensors ρ\rho (indices a,ba,b together with basis indices) and vv (indices i,ji,j together with basis indices), and boundary covectors — satisfying the pentagon-type exchange relation between WW, ρ\rho and vv together with the unit/counit normalization and duality conditions exactly as diagrammed there (these are the axioms under which the associated two-dimensional tensor network has an exact matrix-product-operator representation). Prove or refute: every such finite-dimensional solution is constructible from a finite-dimensional bialgebra A\mathcal{A} together with a representation ρ:A→Matdρ(C)\rho: \mathcal{A} \to \mathrm{Mat}_{d_\rho}(\mathbb{C}) and a dvd_v-dimensional corepresentation {vij∈A}\{v_{ij} \in \mathcal{A}\} via Eq. (2) of the source, namely Wijab=ρab(vij)W^{ab}_{ij} = \rho_{ab}(v_{ij}), ρabx,y=⟨δy⊗ρab,Δ(x)⟩\rho^{x,y}_{ab} = \langle \delta_y \otimes \rho_{ab}, \Delta(x) \rangle, vijx,y=δy(x vij)v^{x,y}_{ij} = \delta_y(x\, v_{ij}) for basis elements x,y∈Ax, y \in \mathcal{A} with dual basis {δx}\{\delta_x\} of A∗\mathcal{A}^*, and boundaries given by the unit and counit of A\mathcal{A}. In the source's words, finite-dimensional bialgebras generate all solutions of Eq. (1). Any resolution must respect the known fact that prebialgebras (algebra-coalgebra structures whose coproduct is multiplicative, Δ(xy)=Δ(x)Δ(y)\Delta(xy) = \Delta(x)\Delta(y), but which need not satisfy the bialgebra compatibility axioms Δ(1)=1⊗1\Delta(1) = 1 \otimes 1 and ϵ(xy)=ϵ(x) ϵ(y)\epsilon(xy) = \epsilon(x)\,\epsilon(y)) solve only a weakened version of the identities.

Component (ii) — the finite-exponent conjecture for C∗C^*-weak Hopf algebras. For any finite-dimensional C∗C^*-weak Hopf algebra A\mathcal{A}, let c=∑x∈B(A)x⊗δx∈A⊗A∗\mathbf{c} = \sum_{x \in B(\mathcal{A})} x \otimes \delta_x \in \mathcal{A} \otimes \mathcal{A}^* be its canonical element. Prove or refute: there exist positive integers η,ν\eta, \nu such that cη+ν=cν\mathbf{c}^{\eta+\nu} = \mathbf{c}^{\nu}. The known positive instance is the Fibonacci/Lee-Yang C∗C^*-weak Hopf algebra, with η=5\eta = 5 and ν=2\nu = 2. If the conjecture holds in general, the revival time of the finite periodic solvable circuits constructed from C∗C^*-weak Hopf algebras is trev=ηLt_{\mathrm{rev}} = \eta L up to divisors, paralleling the C∗C^*-Hopf case cη=1⊗ϵ\mathbf{c}^{\eta} = 1 \otimes \epsilon established via the Etingof-Gelaki exponent theorem.

The common objective of the two components is the algebraic completion of the framework: component (i) decides whether the bialgebra construction is exhaustive (whether it classifies all local gates whose space-time tensor network admits the exact MPO solution mechanism), and component (ii) decides whether the weak-Hopf circuits inherit the finite-exponent, linear-revival structure of the Hopf circuits.

References

  1. Zhiyuan Wang, 'Hopf algebras and solvable unitary circuits', Physical Review B 111, 104315 (2025), DOI 10.1103/PhysRevB.111.104315; arXiv:2409.17215, https://arxiv.org/abs/2409.17215
  2. Pavel Etingof and Shlomo Gelaki, 'On the Exponent of Finite-Dimensional Hopf Algebras', Mathematical Research Letters 6, 131-140 (1999), DOI 10.4310/mrl.1999.v6.n2.a1, https://doi.org/10.4310/mrl.1999.v6.n2.a1
  3. Georgios Styliaris, Rahul Trivedi, and J. Ignacio Cirac, 'Quantum Circuits for Matrix-Product Unitaries', Physical Review Letters 135, 260602 (2025), DOI 10.1103/yshb-hmml; arXiv:2508.08160, https://arxiv.org/abs/2508.08160
  4. He-Ran Wang, Ilya Vilkoviskiy, and Dmitry A. Abanin, 'Temporal Complexity Hierarchies in Solvable Quantum Many-Body Dynamics', arXiv:2510.21927, https://arxiv.org/abs/2510.21927
  5. Bruno Bertini, Pieter W. Claeys, and Tomaž Prosen, 'Exactly solvable quantum many-body dynamics from space-time duality', Reviews of Modern Physics 98, 025001 (2026), DOI 10.1103/yx73-dk86; arXiv:2505.11489, https://arxiv.org/abs/2505.11489
  6. Da-Chuan Lu, Arkya Chatterjee, and Nathanan Tantivasadakarn, 'Generalized Kramers-Wannier Self-Duality in Hopf-Ising Models', arXiv:2602.10183, https://arxiv.org/abs/2602.10183
  7. Yuhan Liu, Alberto Ruiz-de-Alarcón, Georgios Styliaris, Xiao-Qi Sun, David Pérez-García, and J. Ignacio Cirac, 'Parent Lindbladians for matrix product density operators', Physical Review Research 8, 013210 (2026), DOI 10.1103/1qyd-59z7; arXiv:2501.10552, https://arxiv.org/abs/2501.10552
  8. Andras Molnar, Alberto Ruiz-de-Alarcón, José Garre-Rubio, Norbert Schuch, J. Ignacio Cirac, and David Pérez-García, 'Matrix Product Operator Algebras I: Representations of Weak Hopf Algebras and Projected Entangled Pair States', Communications in Mathematical Physics 407, 203 (2026), DOI 10.1007/s00220-026-05670-w; arXiv:2204.05940, https://arxiv.org/abs/2204.05940
  9. I. Colazzo, J. Okniński, and A. Van Antwerpen, 'Bijective solutions to the Pentagon Equation', arXiv:2405.20406, https://arxiv.org/abs/2405.20406
  10. Ilaria Colazzo and Geoffrey Janssens, 'On set-theoretic solutions of pentagon equation and positive basis Hopf algebras', arXiv:2601.22089, https://arxiv.org/abs/2601.22089
  11. Pavel Etingof, Shlomo Gelaki, Dmitri Nikshych, and Victor Ostrik, 'Tensor Categories', Mathematical Surveys and Monographs 205, American Mathematical Society (2015), DOI 10.1090/surv/205, https://doi.org/10.1090/surv/205
  12. Aaron J. Friedman, Sarang Gopalakrishnan, and Romain Vasseur, 'Integrable Many-Body Quantum Floquet-Thouless Pumps', Physical Review Letters 123, 170603 (2019), DOI 10.1103/PhysRevLett.123.170603, https://doi.org/10.1103/PhysRevLett.123.170603
  13. Gabriella Böhm and Kornél Szlachányi, 'A coassociative C*-quantum group with nonintegral dimensions', Letters in Mathematical Physics 38, 437-456 (1996), DOI 10.1007/BF01815526, https://doi.org/10.1007/BF01815526
  14. Richard G. Larson and David E. Radford, 'Semisimple Cosemisimple Hopf Algebras', American Journal of Mathematics 110, 187-195 (1988), DOI 10.2307/2374545, https://doi.org/10.2307/2374545
  15. Kangqiao Li and Shenglin Zhu, 'On the exponent of finite-dimensional non-cosemisimple Hopf algebras', Communications in Algebra 47, 4476-4495 (2019), DOI 10.1080/00927872.2018.1539176, https://doi.org/10.1080/00927872.2018.1539176
  16. Jimin Li, Bruno Bertini, Juan P. Garrahan, and Robert L. Jack, 'Exact large deviations and emergent long-range correlations in sequential quantum East circuits', Physical Review B 113, L100305 (2026), DOI 10.1103/72kx-56lt; arXiv:2509.07182, https://arxiv.org/abs/2509.07182