Classification of unitary Yang–Baxter operators: generating sources, localization of braid representations, and virtual abelianness of braid images

The problem is the classification program for unitary Yang–Baxter operators (unitary R-matrices) proposed by Galindo and Rowell in 2026, together with its two companion conjectures on braid group representations. Conjecture 1.1 asserts that, up to a unit scalar and a local unitary change of basis, every unitary solution of the constant Yang–Baxter equation is generated from three sources — monomial solutions, group-type (Yetter–Drinfeld) solutions, and twisted group-algebra tower solutions — under external products, direct sums, and admissible internal orthogonal sums. Conjecture 1.2 is the Rowell–Wang localization conjecture (2010): the braid representations attached to a simple object of a unitary braided fusion category are localizable by a unitary Yang–Baxter operator exactly when the object has Frobenius–Perron dimension with integral square. Conjecture 1.3 asserts that braid images of unitary Yang–Baxter operators are always virtually abelian. All three conjectures remain open. The source (arXiv:2608.16865) proves, rather than leaves open, the nonemptiness of Lechner's exceptional class $[e^{\pi i/3},1/2,4]$ together with a strict localization of the $SU(3)_3$ Hecke tower, and it verifies an open case ($SO(4)_2$) of the localization conjecture.

Background

A Yang–Baxter operator (or R-matrix) is an invertible linear map $R:V\otimes V\to V\otimes V$ on the tensor square of a finite-dimensional complex Hilbert space $V$ satisfying the constant (quantum) Yang–Baxter equation

$$(R\otimes I)(I\otimes R)(R\otimes I)=(I\otimes R)(R\otimes I)(I\otimes R)$$

in $\mathrm{End}(V\otimes V\otimes V)$. Such an operator yields unitary representations of the braid group $\mathcal{B}_n$ on $V^{\otimes n}$ by sending the Artin generator $\sigma_i$ to the copy of $R$ acting on the $i$-th and $(i+1)$-st tensor factors; when $R$ carries a scalar enhancement, matrix traces on these representations give invariants of oriented links. A unitary Yang–Baxter operator is one with $R$ unitary, so that $R$ is simultaneously a reversible two-particle quantum gate. Classifying all solutions is a classical open problem: the equation is a system of $d^6$ cubic polynomial equations in $d^4$ unknowns ($d=\dim V$), completely solved only for $d=2$ by Hietarinta, and brute-force search is infeasible beyond that. In dimension 2 all unitary solutions are, up to conjugation, monomial (a diagonal unitary times a permutation matrix) with one exception, and all resulting braid images are finite (Rowell–Wang, Section 3). Unitary solutions appear far more constrained than general ones, motivating a conceptual classification. Three classical sources are known. Monomial solutions $R=DP_r$ combine a set-theoretic solution $r$ of the Yang–Baxter equation (a bijection of pairs of labels satisfying the braid relation) with a diagonal phase governed by a multiplicative 2-cocycle. Group-type solutions arise from Yetter–Drinfeld modules over discrete groups (a $G$-graded space with a compatible $G$-action): in a suitable basis $R(x_j\otimes z)=g_j(z)\otimes x_j$ for unitary operators $g_j$. The third source, developed from the iterated twisted tensor-product construction of Gustafson, Kimball, Rowell and Zhang, starts from a finite abelian group carrying a quadratic form (a metric group), a normalized 2-cocycle, a bicharacter, and a coefficient function; these data define braid-compatible unitary elements in a tower of twisted group algebras, and a compatible tensor-power representation turns them into matrix solutions. This source contains the well-studied Gaussian solutions, including the metaplectic examples analyzed by Rowell and Wenzl. Galindo and Rowell conjecture that these three sources, closed under three operations — external (tensor) products, direct sums, and admissible internal orthogonal sums (a replacement that combines two orthogonal partial braid elements inside an already-generated tower) — generate all unitary Yang–Baxter operators up to the equivalence $R\mapsto\lambda(S\otimes S)^{-1}R(S\otimes S)$ with $\lambda\in\mathrm{U}(1)$ and $S\in\mathrm{U}(V)$. The classification is entangled with two older conjectures connecting Yang–Baxter operators to anyon physics. In a unitary braided fusion category (the mathematical model of anyons), a simple object $X$ (one with no nontrivial subobjects) carries braid representations $\rho_n^X$ on $\mathrm{End}(X^{\otimes n})$. These are localizable if, after possibly reassembling their irreducible summands with different positive multiplicities, they are uniformly equivalent, as braid representations, to those of a single unitary Yang–Baxter operator acting on tensor powers $W^{\otimes n}$ — that is, if the anyonic braiding gates can be reassembled into local two-site gates of the circuit model. The Frobenius–Perron dimension $\operatorname{FPdim}(X)$ is the Perron–Frobenius eigenvalue of the fusion matrix of $X$; a category is weakly integral when $\operatorname{FPdim}(X)^2\in\mathbb{Z}$ for all simple $X$. The Rowell–Wang localization conjecture (2010) predicts localizability exactly under weak integrality of $X$; this matters for topological quantum computation because it delineates which anyon models admit local circuit realizations of braiding. Rowell–Wang proved one direction under the hypothesis that the braid image generates the centralizer algebras, implying for instance that the Fibonacci object is not localizable; they also localized the level-2 and level-4 Jones representations, the latter by an explicit unitary $9\times 9$ R-matrix. Galindo and Rowell's virtual-abelianness conjecture (2014) predicts that the braid image $\rho_{R,n}(\mathcal{B}_n)$ of every unitary Yang–Baxter operator is virtually abelian (contains an abelian normal subgroup of finite index) for every $n\geq 2$, so that local unitary braiding never produces a dense nonabelian gate set; it is verified for the Gaussian and group-type families. Adjacent classifications delimit the problem. Lechner, Pennig and Wood classified all involutive unitary R-matrices (spectrum in $\{1,-1\}$) via extremal characters of the infinite symmetric group, the classes being labelled by pairs of Young diagrams. Martin and Rowell classified charge-conserving (spin-chain) solutions, a classification later recast categorically by Martin, Rowell and Torzewska. Lechner (2026) classified unitary R-matrices with exactly two eigenvalues up to equivalence of the full braid-representation tower: all classes lie in eight families, two of which are exhausted by Gaussian R-matrices tensored with identities, leaving one exceptional family $[e^{\pi i/3},1/2,2m]$ of even local dimensions whose existence was open (empty at $d=2$). Galindo and Rowell (2026) settle that existence question: a quaternionic Family III operator, an explicit $16\times 16$ unitary Yang–Baxter operator of local dimension 4, realizes the class $[e^{\pi i/3},1/2,4]$ and strictly localizes the semisimple Hecke braid-image tower of $SU(3)_3$ (the Hecke algebra being the one-parameter quotient of the braid group algebra underlying the HOMFLYPT polynomial); their qutrit (three-level) Family II operator strictly localizes the semisimple quotient of the BMW algebra specializations associated with $SO(4)_2$ (the Birman–Murakami–Wenzl algebra being the two-parameter analogue underlying the Kauffman polynomial), verifying a further open case of the Rowell–Wang conjecture. They also prove finite braid images for all their metric families and virtual abelianness for their dimension-8 cyclic families, even at parameters where the image is infinite. Nevertheless, all three organizing conjectures remain open in general.

Problem Statement

Settle the Galindo–Rowell classification program for unitary Yang–Baxter operators by determining the truth of its three organizing conjectures, in the generality in which the source states them.

(1) Classification conjecture (Conjecture 1.1 of the source, precise form in its Definition 6.1). Let $\mathfrak{G}$ be the smallest class of realized local-tower presentations that contains the braid-image presentations of all monomial and group-type unitary Yang–Baxter operators and all realized presentations arising from unitary braid-compatible coefficient elements in twisted group-algebra towers, and that is closed under external products, direct sums, and admissible internal orthogonal sums (an operation that replaces a braid-compatible element of an already-generated tower by the scaled sum of two orthogonal such elements inside the same tower). Prove or refute: for every unitary Yang–Baxter operator $R\in\mathrm{End}(V\otimes V)$ there exists $\mathfrak{d}\in\mathfrak{G}$ whose local operator satisfies $R_{\mathfrak{d}}=\lambda(S\otimes S)^{-1}R(S\otimes S)$ for some $\lambda\in\mathrm{U}(1)$ and $S\in\mathrm{U}(V)$.

(2) Localization conjecture (Conjecture 1.2 of the source, identical to Conjecture 4.1 of Rowell–Wang). Prove or refute: for every simple object $X$ (an object with no nontrivial subobjects) in a unitary braided fusion category, the sequence of braid group representations $(\rho_n^X,\mathrm{End}(X^{\otimes n}))$ is localizable — that is, uniformly equivalent, as braid representations and up to reassembling the irreducible summands with possibly different positive multiplicities (Definition 2.3 of Rowell–Wang), to the tensor-power representations of a single unitary Yang–Baxter operator — if and only if $\operatorname{FPdim}(X)^2\in\mathbb{Z}$, where $\operatorname{FPdim}(X)$ is the Frobenius–Perron dimension of $X$ (the Perron–Frobenius eigenvalue of the fusion matrix of $X$).

(3) Virtual-abelianness conjecture (Conjecture 1.3 of the source, identical to Conjecture 1.1 of Galindo–Rowell 2014). Prove or refute: for every unitary Yang–Baxter operator $R$ and every $n\geq 2$, the braid image $\rho_{R,n}(\mathcal{B}_n)\subset\mathrm{U}(V^{\otimes n})$ is virtually abelian (contains an abelian normal subgroup of finite index).

A complete answer settles each of (1)–(3) over all local dimensions $d\geq 2$ (and, for (2), for all simple objects of unitary braided fusion categories), by proof or by explicit counterexample; an answer settling a proper subset is partial progress and is evaluated per conjecture. The three conjectures form one scientific objective: (1) describes all unitary Yang–Baxter operators, (2) describes exactly which categorical (anyon-model) braid representations they can realize, and (3) describes the braid images they produce; (1) is the conjectural route to (3), since finiteness and virtual abelianness are proved for the basic metric and cyclic families but not for arbitrary elements of the closure $\mathfrak{G}$. The sub-tasks flagged in the source's outlook — characteristic-zero versions of the ququint gauge comparisons, an intrinsic theory of the admissible internal orthogonal sum, and searches beyond the Pauli and Clifford ansätze — are natural components of a proof strategy, not separately required deliverables.

References

1. C. Galindo and E. C. Rowell, 'Unitary Yang--Baxter Operators: Towards a Classification', arXiv:2608.16865 (2026), doi:10.48550/arXiv.2608.16865, https://arxiv.org/abs/2608.16865
2. E. C. Rowell and Z. Wang, 'Localization of unitary braid group representations', Communications in Mathematical Physics 311 (2012), no. 3, 595-615, doi:10.1007/s00220-011-1386-7, arXiv:1009.0241, https://arxiv.org/abs/1009.0241
3. C. Galindo and E. C. Rowell, 'Braid representations from unitary braided vector spaces', Journal of Mathematical Physics 55 (2014), no. 6, 061702, doi:10.1063/1.4880196, arXiv:1312.5557, https://arxiv.org/abs/1312.5557
4. C. Galindo, S.-M. Hong, and E. C. Rowell, 'Generalized and quasi-localizations of braid group representations', International Mathematics Research Notices 2013 (2013), no. 3, 693-731, doi:10.1093/imrn/rnr269, arXiv:1105.5048, https://arxiv.org/abs/1105.5048
5. G. Lechner, 'The classification problem for unitary R-Matrices with two eigenvalues', arXiv:2603.20158 (2026), doi:10.48550/arXiv.2603.20158, https://arxiv.org/abs/2603.20158
6. G. Lechner, U. Pennig, and S. Wood, 'Yang-Baxter representations of the infinite symmetric group', Advances in Mathematics 355 (2019), 106769, doi:10.1016/j.aim.2019.106769, arXiv:1707.00196, https://arxiv.org/abs/1707.00196
7. P. P. Martin and E. C. Rowell, 'Classification of spin-chain braid representations', Communications in Mathematical Physics 407 (2026), no. 10, article 201, doi:10.1007/s00220-026-05716-z, arXiv:2112.04533, https://arxiv.org/abs/2112.04533
8. P. P. Martin, E. C. Rowell, and F. Torzewska, 'A Categorical Perspective on Braid Representations', arXiv:2506.07950 (2025), https://arxiv.org/abs/2506.07950
9. E. C. Rowell, 'A quaternionic braid representation (after Goldschmidt and Jones)', Quantum Topology 2 (2011), no. 2, 173-182, doi:10.4171/qt/18, arXiv:1006.4808, https://arxiv.org/abs/1006.4808
10. E. C. Rowell and H. Wenzl, 'SO(N)_2 Braid group representations are Gaussian', Quantum Topology 8 (2017), no. 1, 1-33, doi:10.4171/qt/85, arXiv:1401.5329, https://arxiv.org/abs/1401.5329
11. P. Gustafson, A. Kimball, E. C. Rowell, and Q. Zhang, 'Braid group representations from twisted tensor products of algebras', Peking Mathematical Journal 3 (2020), no. 2, 103-130, doi:10.1007/s42543-020-00023-5, arXiv:1906.08153, https://arxiv.org/abs/1906.08153
12. J. Hietarinta, 'All solutions to the constant quantum Yang-Baxter equation in two dimensions', Physics Letters A 165 (1992), no. 3, 245-251, doi:10.1016/0375-9601(92)90044-M, https://doi.org/10.1016/0375-9601(92)90044-M
13. R. Conti and G. Lechner, 'Yang–Baxter endomorphisms', Journal of the London Mathematical Society 103 (2021), no. 2, 633-671, doi:10.1112/jlms.12387, arXiv:1909.04127, https://arxiv.org/abs/1909.04127
14. Y. S. Baki and P. Veerapen, 'Gaussian solutions to the Yang--Baxter equation and their twists', arXiv:2607.13876 (2026), https://arxiv.org/abs/2607.13876