Modified operators and off-shell relations beyond the rational six-vertex model: the trigonometric anti-periodic, reflection-algebra, and higher-rank extensions of the linear-system method

The linear-system method of the modified algebraic Bethe ansatz derives determinant formulas by showing that a quantity of interest — a scalar product of Bethe states or a lattice partition function — satisfies a homogeneous linear system whose solution is unique up to normalization and is a determinant. Belliard, Pimenta and Slavnov (SciPost Phys. 16, 009 (2024)) used it to express the partition function of the rational six-vertex model on a rectangle with generalized domain-wall-type boundary conditions through a modified Izergin determinant, and stated three extensions for which the construction does not yet exist: the trigonometric (XXZ) six-vertex model with anti-periodic boundary conditions, for which the required modified operators and off-shell relations "are not yet known"; reflecting boundary conditions governed by the reflection algebra; and higher-rank models based on sl(n). No construction of the missing objects in any of the three settings is known in the literature to date; adjacent separation-of-variables, functional-relation, and Izergin-type results address different quantities by different methods. The problem — construct the modified operators and off-shell triangular relations in these settings and derive the corresponding linear systems with determinant-type solutions — remains open.

Background

The six-vertex model assigns arrows to the edges of a square lattice under the ice rule: at each vertex exactly two of the four incident arrows point inward. On an $m\times n$ rectangle with domain wall boundary conditions (DWBC — all boundary arrows point inward on two opposite sides and outward on the other two; the allowed configurations are in bijection with alternating sign matrices), the inhomogeneous partition function is given by a single determinant, the Izergin–Korepin determinant, a cornerstone of exactly solved two-dimensional statistical mechanics since the 1980s (Korepin, 1982; the full lineage is reviewed in the source). The quantum inverse scattering method (QISM) reformulates such partition functions algebraically. A monodromy matrix $T(u)$ — a $2\times2$ matrix whose entries are operators — encodes the model through the Yang–Baxter ($RTT$) commutation relations with the rational $R$-matrix (the XXX chain, governed by the Yangian $Y(\mathrm{gl}_2)$, the algebra generated by the monodromy entries) or the trigonometric $R$-matrix (the XXZ chain, governed by the quantum group $U_q(\widehat{\mathfrak{sl}}_2)$). In the algebraic Bethe ansatz (ABA), eigenstates of the transfer matrix $t(u)=\operatorname{tr}T(u)$, or of its twisted version $t(u)=\operatorname{tr}\,K\,T(u)$ with a boundary twist matrix $K$, are built by acting with creation operators on a reference state (pseudo-vacuum), and scalar products and norms of Bethe states then admit determinant representations. This construction requires a $U(1)$ symmetry: $K$ must be diagonal so that the total spin is conserved and a reference state exists. Anti-periodic boundary conditions correspond to a non-diagonal twist (for instance $K\propto\sigma_x$, a Pauli matrix), which breaks the $U(1)$ symmetry and makes ordinary ABA fail. The modified algebraic Bethe ansatz (MABA) was developed for such models without $U(1)$ symmetry: one introduces modified creation operators — concrete combinations of monodromy entries, or operators built from a second, gauge-transformed monodromy matrix — whose action on suitable off-shell (non-eigenstate) vectors is still triangular. The central technical objects are off-shell (triangular) relations: identities giving the action of the monodromy entries on strings of modified operators carrying arbitrary, non-solution ("off-shell") spectral parameters. Belliard and Slavnov observed that such off-shell relations force a scalar product to satisfy a homogeneous linear system in its variables whose solution is unique up to normalization and is a determinant — the linear-system approach. It has been carried out for on-shell/off-shell scalar products in the twisted XXX chain (the modified algebraic Bethe ansatz for that chain is due to Belliard–Slavnov–Vallet, 2018; the linear-system route to scalar products originates with Belliard–Slavnov, 2019), in the 8-vertex (XYZ) chain (Slavnov–Zabrodin–Zotov, 2020), and in the open XXZ chain with general integrable boundaries (Belliard–Pimenta–Slavnov, 2021). Belliard–Pimenta–Slavnov (2024) extended the method to partition functions. It considers the rational six-vertex model on an $m\times n$ rectangle with boundary conditions generalizing DWBC: four arbitrary boundary walls ("compass states"), described by rank-one boundary matrices or twists, two on the vertical and two on the horizontal direction. Using the triangular representation theory of the modified operators in the Yangian, the paper derives a linear system satisfied by the partition function and solves it by a modified Izergin determinant that recovers, at special parameter values, both the partial domain wall partition function and the Izergin–Korepin determinant itself. Its conclusion names three directions in which the analogous construction is missing. First, the trigonometric six-vertex model (the XXZ chain) with anti-periodic boundary conditions: the appropriate modified operators and off-shell relations "are not yet known for this model". The gap is not due to a lack of alternatives: separation of variables (SoV — a method that constructs eigenstates from a separated coordinate basis rather than from creation operators acting on a reference state) yields the complete spectrum of antiperiodic XXZ chains (Niccoli–Terras, 2015) and even determinant representations for their scalar products and form factors (Pei–Terras, 2022), and functional $t$–$W$ relations give the thermodynamics at the special anisotropy $\eta=i\pi/3$ (Lu–Cao–Yang–Marquette–Zhang, 2024). But none of this provides the ABA-type modified operators, the off-shell Bethe vectors, or the rectangular generalized-DWBC partition function at which the linear-system method aims. Second, reflecting boundary conditions, governed by the reflection algebra (the boundary analogue of the $RTT$ algebra introduced by Sklyanin): determinant formulas for several reflecting-end domain wall partition functions exist through the original Izergin-type methods (Tsuchiya, 1998, diagonal reflecting end; Yang–Chen–Feng–Hao–Hou–Shi–Zhang, 2011, non-diagonal reflecting end; Hietala, 2022, partially reflecting end) and through half-space symmetric-function and Pfaffian methods (Garbali–de Gier–Mead–Wheeler, 2025), but not through the linear-system method or the generalized four-wall boundary geometry. Third, higher-rank vertex models based on $\mathfrak{sl}(n)$: nested ABA yields Bethe vectors and commutation relations, and a nested Izergin–Korepin analysis characterizes certain higher-rank partition functions (Gerrard–Motegi–Sakai, 2025), but the modified-operator off-shell relations and the associated linear systems are unconstructed.

Problem Statement

Extend the modified-operator/linear-system method realized for the rational six-vertex model by Belliard–Pimenta–Slavnov (2024) beyond the rational case, by carrying out in each of the following settings the same four-step program: (i) construct the appropriate modified creation operators (and their companion monodromy-type operators) in the algebraic framework of the setting; (ii) prove the off-shell triangular relations — the action of the algebra's generators on arbitrary strings of the modified operators, valid for arbitrary (off-shell) spectral parameters; (iii) derive the homogeneous linear system that characterizes, up to normalization, the corresponding partition function on the rectangular lattice with the generalized domain-wall-type boundary conditions, including the argument that the system determines its solution; (iv) solve the linear system in explicit determinant-type form, as in the rational prototype's modified Izergin determinant. The three settings are: Setting A (explicitly open in the source): the trigonometric six-vertex model (XXZ chain, quantum-group $R$-matrix) with anti-periodic boundary conditions, for which the source states the appropriate modified operators and off-shell relations "are not yet known", identifying their construction as the first step toward determinant formulas for that model. Setting B: the six-vertex model with reflecting boundary conditions, governed by the reflection algebra — the linear-system derivation of rectangular partition functions with reflecting ends under the generalized boundary geometry, for which determinant formulas currently exist only through other methods. Setting C: higher-rank vertex models based on $\mathfrak{sl}(n)$, whose off-shell relations and associated linear systems the source describes as considerably more intricate than the rank-one case. The source poses the three settings as separate open directions of one program — extending the modified-operator route to linear systems and determinant formulas beyond the rational case: Setting A is the explicitly open problem, its modified operators and off-shell relations being "not yet known" and their construction the first step toward determinant formulas for that model; Setting B is the direction the source raises as interesting to consider, and Setting C the one it describes as important to investigate, with off-shell relations and associated linear systems more intricate than in the rank-one case. The program above is required in all three settings; the source's differing emphasis on them — explicitly posed for Setting A, raised as interesting to consider for Setting B, described as important to investigate for Setting C — reflects its framing of the directions and does not make any of them optional. Constructions must be explicit (operators given as concrete combinations of monodromy-matrix entries, double-row/reflection monodromies, or nested monodromies). In the rational limit $q\to1$, the Setting A model reduces to the rational six-vertex model treated in the source; Settings B and C build on the established reflection-algebra and nested algebraic-Bethe-ansatz frameworks.

References

1. S. Belliard, R. A. Pimenta, N. A. Slavnov, 'Modified rational six vertex model on the rectangular lattice', SciPost Phys. 16, 009 (2024). DOI: 10.21468/SciPostPhys.16.1.009; arXiv:2310.05850, https://arxiv.org/abs/2310.05850
2. M. Cornillault, S. Belliard, 'Modified rational six vertex model on a rectangular lattice: new formula, homogeneous and thermodynamic limits', SciPost Phys. 20, 076 (2026). DOI: 10.21468/SciPostPhys.20.3.076; arXiv:2509.11797, https://arxiv.org/abs/2509.11797
3. P. Baseilhac, R. A. Pimenta, 'The q-Racah polynomials from scalar products of Bethe states II', J. Phys. A: Math. Theor. 58, 125205 (2025). DOI: 10.1088/1751-8121/adbfe7; arXiv:2501.10310, https://arxiv.org/abs/2501.10310
4. K. Motegi, 'Factorization of rational six vertex model partition functions', Nucl. Phys. B 1009, 116743 (2024). DOI: 10.1016/j.nuclphysb.2024.116743; arXiv:2410.03153, https://arxiv.org/abs/2410.03153
5. K. Motegi, R. Ohkawa, 'Algebraic formulas and geometric derivation of source identities', J. Phys. A: Math. Theor. 57, 325201 (2024). DOI: 10.1088/1751-8121/ad62c8; arXiv:2404.08141, https://arxiv.org/abs/2404.08141
6. S. Belliard, A. Hutsalyuk, 'Full counting statistics for twisted XXX spin chains' (2025). arXiv:2509.13928, https://arxiv.org/abs/2509.13928
7. S. Belliard, R. A. Pimenta, N. A. Slavnov, 'Scalar product for the XXZ spin chain with general integrable boundaries', J. Phys. A: Math. Theor. 54, 344001 (2021). DOI: 10.1088/1751-8121/ac1482; arXiv:2103.12501, https://arxiv.org/abs/2103.12501
8. S. Belliard, N. A. Slavnov, B. Vallet, 'Modified Algebraic Bethe Ansatz: Twisted XXX Case', SIGMA 14, 054 (2018). DOI: 10.3842/SIGMA.2018.054; arXiv:1804.00597, https://arxiv.org/abs/1804.00597
9. S. Belliard, N. A. Slavnov, 'Why scalar products in the algebraic Bethe ansatz have determinant representation', JHEP 10, 103 (2019). DOI: 10.1007/JHEP10(2019)103; arXiv:1908.00032, https://arxiv.org/abs/1908.00032
10. N. Slavnov, A. Zabrodin, A. Zotov, 'Scalar products of Bethe vectors in the 8-vertex model', JHEP 06 (2020) 123. DOI: 10.1007/JHEP06(2020)123; arXiv:2005.11224, https://arxiv.org/abs/2005.11224
11. H. Pei, V. Terras, 'On scalar products and form factors by separation of variables: the antiperiodic XXZ model', J. Phys. A: Math. Theor. 55, 015205 (2022). DOI: 10.1088/1751-8121/ac3b85; arXiv:2011.06109, https://arxiv.org/abs/2011.06109
12. G. Niccoli, V. Terras, 'Antiperiodic XXZ chains with arbitrary spins: Complete eigenstate construction by functional equations in separation of variables', Lett. Math. Phys. 105, 989 (2015). DOI: 10.1007/s11005-015-0759-9; arXiv:1411.6488, https://arxiv.org/abs/1411.6488
13. P. Lu, J. Cao, W.-L. Yang, I. Marquette, Y.-Z. Zhang, 'T-W relation and free energy of the antiperiodic XXZ chain with η=iπ/3 at a finite temperature', J. Phys. A: Math. Theor. 57, 445205 (2024). DOI: 10.1088/1751-8121/ad85b3; arXiv:2402.04849, https://arxiv.org/abs/2402.04849
14. O. Tsuchiya, 'Determinant formula for the six-vertex model with reflecting end', J. Math. Phys. 39, 5946 (1998). DOI: 10.1063/1.532606, https://doi.org/10.1063/1.532606
15. W.-L. Yang, X. Chen, J. Feng, K. Hao, B.-Y. Hou, K.-J. Shi, Y.-Z. Zhang, 'Determinant formula for the partition function of the six-vertex model with a non-diagonal reflecting end', Nucl. Phys. B 844, 289 (2011). DOI: 10.1016/j.nuclphysb.2010.11.008, https://doi.org/10.1016/j.nuclphysb.2010.11.008
16. L. Hietala, 'Exact results for the six-vertex model with domain wall boundary conditions and a partially reflecting end', Lett. Math. Phys. 112, 41 (2022). DOI: 10.1007/s11005-022-01530-5; arXiv:2104.05389, https://arxiv.org/abs/2104.05389
17. A. Garbali, J. de Gier, W. Mead, M. Wheeler, 'Symmetric functions from the six-vertex model in half-space', Ann. Henri Poincaré 26, 2557 (2025). DOI: 10.1007/s00023-024-01484-5; arXiv:2312.14348, https://arxiv.org/abs/2312.14348
18. A. J. Gerrard, K. Motegi, K. Sakai, 'Higher rank elliptic partition functions and multisymmetric elliptic functions', Nucl. Phys. B 1011, 116805 (2025). DOI: 10.1016/j.nuclphysb.2025.116805; arXiv:2412.13561, https://arxiv.org/abs/2412.13561
19. E. K. Sklyanin, 'Boundary conditions for integrable quantum systems', J. Phys. A: Math. Gen. 21, 2375 (1988). DOI: 10.1088/0305-4470/21/10/015, https://doi.org/10.1088/0305-4470/21/10/015
20. V. E. Korepin, 'Calculation of norms of Bethe wave functions', Comm. Math. Phys. 86, 391 (1982). DOI: 10.1007/BF01212176, https://doi.org/10.1007/BF01212176