The open asymmetric simple exclusion process (ASEP) — a chain of NN sites coupled to particle reservoirs at both ends — is the paradigmatic exactly tractable model of a boundary-driven nonequilibrium steady state, and the full counting statistics of the current exchanged with a reservoir — encoded in the scaled cumulant generating function E(μ)E(\mu), the largest eigenvalue Λ0\Lambda_0 of the current-deformed Markov matrix M(ξ)M(\xi) at fugacity ξ=eμ\xi=e^{\mu} — is its central still-unresolved observable at finite size and generic parameters. Finn and Vanicat (2017) solved, in matrix-product form, boundary quantum Knizhnik–Zamolodchikov (qKZ) equations with an additional deformation, at the constraint ξ=sm\xi=s^{m}, and proved that the normalization Z(m)Z^{(m)} of their twice-deformed ground state is the symmetric Koornwinder polynomial P(mN)P_{(m^{N})} with rectangular partition. They conjectured (Conjecture 4.1) that, as mm tends to infinity with s=ξ1/ms=\xi^{1/m}, the normalized states converge to a regular vector ∣Ψ0⟩|\Psi_{0}\rangle and ln⁡ξmln⁡Z(m)\frac{\ln\xi}{m}\ln Z^{(m)} converges to a regular function F0F_{0}, from which the current statistics follows through E(μ)=p−q2 ∂2F0∂xi2(1;eμ)E(\mu)=\frac{p-q}{2}\,\frac{\partial^{2}F_{0}}{\partial x_i^{2}}(\mathbf{1};e^{\mu}) evaluated at 1=(1,…,1)\mathbf{1}=(1,\dots,1). The conjecture is supported only by numerics up to three sites; the characterization of F0F_{0} by a nonlinear functional equation has no known closed-form or constructive solution, so the formula for E(μ)E(\mu) remains formal. The problem is to prove or disprove the conjectured convergence and — should it hold — to determine F0F_{0} effectively, thereby making the exact current statistics of the finite open ASEP computable non-perturbatively.

Background

The asymmetric simple exclusion process (ASEP) is a stochastic lattice gas on a chain of NN sites in which particles hop right and left with rates pp and qq, with at most one particle per site (exclusion). In the open ASEP the chain is coupled to two reservoirs: particles are injected at the left end with rate α\alpha and removed with rate γ\gamma; at the right end, particles are extracted with rate β\beta and injected with rate δ\delta (the boundary-rate convention of the source paper). The system relaxes to a nonequilibrium steady state carrying a macroscopic current, and its exact stationary measure is given by a matrix product ansatz, in which the stationary weight of a configuration is a product of non-commuting matrices obeying quadratic algebraic relations (Derrida, Evans, Hakim and Pasquier, 1993).

Because the current is the order parameter of a boundary-driven nonequilibrium system, its fluctuation statistics is a central observable. Writing QTQ_T for the algebraic number of particles exchanged between the left reservoir and the chain up to time TT, the probability of an atypical flux obeys P(QT/T=j)∼e−TG(j)P(Q_T/T=j)\sim e^{-T G(j)} at large TT, where G(j)G(j) is the current large deviation function, the Legendre transform of the scaled cumulant generating function E(μ)E(\mu). The latter is obtained by deforming the Markov generator by a fugacity ξ=eμ\xi=e^{\mu} counting each jump across the left boundary, M(ξ)=αγ B1(ξ)+∑i=1N−1pq wi,i+1+βδ BNM(\xi)=\sqrt{\alpha\gamma}\,B_1(\xi)+\sum_{i=1}^{N-1}\sqrt{pq}\,w_{i,i+1}+\sqrt{\beta\delta}\,B_N (the local bulk matrix ww and boundary matrices B1(ξ),BNB_1(\xi),B_N encode the hopping and reservoir rates; M(1)M(1) is the true stochastic generator), and taking its largest eigenvalue: E(μ)=Λ0(eμ)E(\mu)=\Lambda_0(e^{\mu}), with Λ0(1)=0\Lambda_0(1)=0. The Gallavotti–Cohen fluctuation symmetry relates large positive and negative currents; for the open ASEP it states that Λ0\Lambda_0 is invariant under ξ→ξ′=(γδ/αβ)(q/p)N−1 ξ−1\xi\to\xi'=(\gamma\delta/\alpha\beta)(q/p)^{N-1}\,\xi^{-1}.

What is known about E(μ)E(\mu): for the periodic ASEP the current large deviation function is exactly known (Derrida and Lebowitz, 1998); for the open TASEP (the totally asymmetric case q=0q=0) an exact formula for the statistics of the current exists (Lazarescu and Mallick, 2011); for the open ASEP in the thermodynamic limit, de Gier and Essler (2011) used the Bethe ansatz — the exact method that represents eigenvectors as superpositions of plane waves, here applied to the deformed stochastic generator — to obtain the current statistics at the first site in the low- and high-density phases, and the exact diffusion constant in the steady state is known for all NN (Derrida, Evans and Mallick, 1995); at finite size, the perturbative matrix ansatz (Gorissen, Lazarescu, Mallick and Vanderzande, 2012; Lazarescu, 2013) yields every cumulant of the current order by order in the expansion about equilibrium ξ=1\xi=1, at the price of solving nonlinear implicit equations at each order, and Lazarescu and Pasquier (2014) constructed for the deformed open ASEP a Baxter QQ-operator — an auxiliary operator commuting with the deformed transfer matrix, a standard route to its spectrum — and the associated tt-QQ relations. A closed, non-perturbative expression for E(μ)E(\mu) at finite NN and generic parameters is nevertheless not known, a gap explicitly recorded in Lazarescu's 2015 review of current fluctuations in bulk-driven lattice gases.

Finn and Vanicat (2017) connected this problem to the algebraic combinatorics of the model. The stationary states of multispecies open ASEP are known to be tied to multivariate Koornwinder polynomials (Cantini, 2015; Cantini, Garbali, de Gier and Wheeler, 2016) — the Macdonald polynomials of affine type CC, a multiparameter family of symmetric Laurent polynomials generalizing the one-variable Askey–Wilson polynomials (Cantini, de Gier and Wheeler, 2015, established the analogous Macdonald result for the periodic chain). The symmetric Koornwinder polynomial PλP_\lambda with partition λ\lambda is the unique W0W_0-invariant element of its span with unit leading monomial, where W0W_0 is the finite Weyl group of type CNC_N generated by the transpositions si:xi↔xi+1s_i:x_i\leftrightarrow x_{i+1} (1≤i≤N−11\le i\le N-1) and the inversion sN:xN↦xN−1s_N:x_N\mapsto x_N^{-1}, i.e. the signed permutations of the variables x1,…,xNx_1,\dots,x_N; it is an eigenfunction, with eigenvalue dλd_\lambda, of the conserved difference operator

D=∑i=1N[gi(x)(Ts,i−1)+gi(x−1)(Ts,i−1−1)],D=\sum_{i=1}^{N}\Big[g_i(\mathbf{x})(T_{s,i}-1)+g_i(\mathbf{x}^{-1})(T_{s,i}^{-1}-1)\Big],

where Ts,iT_{s,i} is the shift xi↦s xix_i\mapsto s\,x_i and the gig_i are explicit rational functions of x\mathbf{x} depending on the Koornwinder parameters. The ASEP rates fix these parameters through the dictionary t=p/qt=p/q, t0=α/γt_0=\alpha/\gamma, tN=β/δt_N=\beta/\delta, with u0,uNu_0,u_N determined by u01/2−u0−1/2=(p−q+γ−α)/αγu_0^{1/2}-u_0^{-1/2}=(p-q+\gamma-\alpha)/\sqrt{\alpha\gamma} and uN1/2−uN−1/2=(p−q+δ−β)/βδu_N^{1/2}-u_N^{-1/2}=(p-q+\delta-\beta)/\sqrt{\beta\delta}. Finn and Vanicat introduced a further (second) deformation of the current-deformed ground state by solving boundary quantum Knizhnik–Zamolodchikov (qKZ) equations — difference equations, with step ss in the spectral variables, satisfied by correlation functions of integrable models — in matrix-product form in an infinite-dimensional Fock-space representation of the boundary Yang–Baxter algebra, the algebra generated by the bulk and boundary matrices of the underlying integrable open-chain structure. At the constraint ξ=sm\xi=s^m (mm a positive integer) they proved (Theorem 3.14) that the normalization Z(m)(x;s)=⟨1∣Ψ(m)(x;s)⟩Z^{(m)}(\mathbf{x};s)=\langle 1|\Psi^{(m)}(\mathbf{x};s)\rangle, i.e. the sum of components of the twice-deformed ground state, equals exactly the symmetric Koornwinder polynomial P(mN)(x)P_{(m^N)}(\mathbf{x}) with rectangular partition (m,…,m)(m,\dots,m).

The link to current statistics arises in the simultaneous limit m→∞m\to\infty, s=ξ1/m→1s=\xi^{1/m}\to1, a quasi-classical-type degeneration of the qKZ equations (a limit in which the step parameter tends to 11 and difference equations degenerate towards differential equations), analogous to the quasi-classical asymptotics of Knizhnik–Zamolodchikov-type equations studied by Reshetikhin (1992) and Reshetikhin and Varchenko (1994). Finn and Vanicat conjecture (Conjecture 4.1), on the basis of numerics up to N=3N=3 sites, that in this limit the normalized states converge to a vector ∣Ψ0(x;ξ)⟩|\Psi_0(\mathbf{x};\xi)\rangle regular in x\mathbf{x} and that ln⁡ξmln⁡Z(m)(x; s=ξ1/m)\frac{\ln\xi}{m}\ln Z^{(m)}(\mathbf{x};\,s=\xi^{1/m}) converges to a function F0(x;ξ)F_0(\mathbf{x};\xi) regular in x\mathbf{x}. Treating the conjecture as fact, they then derive (Propositions 4.2–4.5): F0F_0 is W0W_0-invariant with the Gallavotti–Cohen symmetry on its gradient; ∣Ψ0⟩|\Psi_0\rangle is an eigenvector of the qKZ scattering matrices at s=1s=1 with eigenvalue exp⁡(xi ∂F0/∂xi)\exp(x_i\,\partial F_0/\partial x_i); the ground-state relation of the deformed Markov matrix,

M(ξ) ∣Ψ0(1;ξ)⟩=p−q2 ∂2F0∂xi2(1;ξ) ∣Ψ0(1;ξ)⟩,M(\xi)\,|\Psi_0(\mathbf{1};\xi)\rangle=\frac{p-q}{2}\,\frac{\partial^2 F_0}{\partial x_i^2}(\mathbf{1};\xi)\,|\Psi_0(\mathbf{1};\xi)\rangle,

which yields the generating function of the cumulants of the current as E(μ)=Λ0(eμ)=p−q2∂2F0∂xi2(1; eμ)E(\mu)=\Lambda_0(e^{\mu})=\frac{p-q}{2}\frac{\partial^2 F_0}{\partial x_i^2}(\mathbf{1};\,e^{\mu}) with 1=(1,…,1)\mathbf{1}=(1,\dots,1) (the right-hand side being independent of ii by W0W_0 invariance); and a characterization of F0F_0 as the solution of the nonlinear functional equation

∑i=1N[gi(x;1)(exi ∂F0/∂xi−1)+gi(x−1;1)(e−xi ∂F0/∂xi−1)]=d0(ξ),\sum_{i=1}^{N}\Big[g_i(\mathbf{x};1)\big(e^{x_i\,\partial F_0/\partial x_i}-1\big)+g_i(\mathbf{x}^{-1};1)\big(e^{-x_i\,\partial F_0/\partial x_i}-1\big)\Big]=d_0(\xi),

obtained as the m→∞m\to\infty limit of the Koornwinder eigenvalue equation DP(mN)=d(mN)P(mN)DP_{(m^N)}=d_{(m^N)}P_{(m^N)}, with d0(ξ)=1−tN1−t(ξ−1)(t0tNtN−1−ξ−1)d_0(\xi)=\frac{1-t^N}{1-t}(\xi-1)\big(t_0t_Nt^{N-1}-\xi^{-1}\big). The authors stress that in practice this does not yet compute the current cumulants, because no closed expression for F0F_0 is known. They also record two adjacent questions: whether the characterization of F0F_0 can be reformulated through Baxter tt-QQ relations (with the stationary normalization of inhomogeneous discrete-time exclusion processes, which satisfies such a difference equation, and the Q-operator of Lazarescu and Pasquier as precedents), and whether the TASEP limit t→∞t\to\infty (i.e. q→0q\to0) of the construction simplifies it.

Problem Statement

For the open ASEP with NN sites, generic bulk rates p,q>0p,q>0 with p≠qp\ne q and generic boundary rates α,β,γ,δ>0\alpha,\beta,\gamma,\delta>0, let M(ξ)M(\xi) be the current-counting deformed Markov matrix above, and let ∣Ψ(m)(x;s)⟩|\Psi^{(m)}(\mathbf{x};s)\rangle be Finn and Vanicat's twice-deformed ground state: the matrix-product solution of the boundary qKZ equations at the constraint ξ=sm\xi=s^m (mm a positive integer), with normalization Z(m)(x;s)=⟨1∣Ψ(m)(x;s)⟩=P(mN)(x)Z^{(m)}(\mathbf{x};s)=\langle1|\Psi^{(m)}(\mathbf{x};s)\rangle=P_{(m^N)}(\mathbf{x}), the symmetric Koornwinder polynomial with rectangular partition (m,…,m)(m,\dots,m) in the ASEP parameter dictionary (t=p/qt=p/q, t0=α/γt_0=\alpha/\gamma, tN=β/δt_N=\beta/\delta, u0,uNu_0,u_N as above).

(i) Prove, or disprove, Conjecture 4.1 of Finn and Vanicat: as m→∞m\to\infty with s=ξ1/ms=\xi^{1/m},

lim⁡m→∞∣Ψ(m)(x; s=ξ1/m)⟩Z(m)(x; s=ξ1/m)=∣Ψ0(x;ξ)⟩,\lim_{m\to\infty}\frac{|\Psi^{(m)}(\mathbf{x};\,s=\xi^{1/m})\rangle}{Z^{(m)}(\mathbf{x};\,s=\xi^{1/m})}=|\Psi_0(\mathbf{x};\xi)\rangle, lim⁡m→∞ln⁡ξm ln⁡Z(m)(x; s=ξ1/m)=lim⁡m→∞ln⁡ξm ln⁡P(mN)(x; s=ξ1/m)=F0(x;ξ),\lim_{m\to\infty}\frac{\ln\xi}{m}\,\ln Z^{(m)}(\mathbf{x};\,s=\xi^{1/m})=\lim_{m\to\infty}\frac{\ln\xi}{m}\,\ln P_{(m^N)}(\mathbf{x};\,s=\xi^{1/m})=F_0(\mathbf{x};\xi),

with ∣Ψ0(x;ξ)⟩|\Psi_0(\mathbf{x};\xi)\rangle and F0(x;ξ)F_0(\mathbf{x};\xi) regular functions of x=(x1,…,xN)\mathbf{x}=(x_1,\dots,x_N) (regular meaning analytic, i.e. holomorphic, in x\mathbf{x}).

(ii) Assuming the convergence in (i), determine F0F_0 effectively for general NN and generic parameters, as a regular W0W_0-invariant solution of the functional equation

∑i=1N[gi(x;1)(exi ∂F0/∂xi−1)+gi(x−1;1)(e−xi ∂F0/∂xi−1)]=d0(ξ),\sum_{i=1}^{N}\Big[g_i(\mathbf{x};1)\big(e^{x_i\,\partial F_0/\partial x_i}-1\big)+g_i(\mathbf{x}^{-1};1)\big(e^{-x_i\,\partial F_0/\partial x_i}-1\big)\Big]=d_0(\xi),

with d0(ξ)=1−tN1−t(ξ−1)(t0tNtN−1−ξ−1)d_0(\xi)=\frac{1-t^N}{1-t}(\xi-1)(t_0t_Nt^{N-1}-\xi^{-1}) and gig_i the rational coefficients of the Koornwinder difference operator DD defined in the background, so that the relation

E(μ)=Λ0(eμ)=p−q2 ∂2F0∂xi2(1; eμ)E(\mu)=\Lambda_0(e^{\mu})=\frac{p-q}{2}\,\frac{\partial^2F_0}{\partial x_i^2}(\mathbf{1};\,e^{\mu})

becomes an effective, non-perturbative, exact formula for the scaled cumulant generating function of the current of the finite open ASEP, from which all current cumulants and the current large deviation function follow.

The core open question is the convergence in (i). A disproof of (i) — non-existence or non-regularity of the limits for an admissible parameter set — settles the conjecture negatively and, with it, the conditional status of Propositions 4.2–4.5 of the source; part (ii) then becomes void, since F0F_0 and the functional equation of Proposition 4.5 are defined only under the conjecture. Part (ii) reflects the source's remark that, even granting the conjecture, the formula for E(μ)E(\mu) 'does not help to compute the cumulants of the current because a closed expression for F0F_0 is missing'; it is a follow-on goal conditional on (i), and the source states the functional equation only as a characterization of F0F_0, attaching no uniqueness claim and no completion criterion to it. Two sub-directions flagged by the authors fall within the same objective but do not separately bound completion: whether the functional equation for F0F_0 can be recast as a Baxter tt-QQ relation, and whether the TASEP limit t→∞t\to\infty (i.e. q→0q\to0) of the construction simplifies it.

References

  1. C. Finn and M. Vanicat, 'Matrix product construction for Koornwinder polynomials and fluctuations of the current in the open ASEP', J. Stat. Mech. Theory Exp. 2017, 023102 (2017). DOI: 10.1088/1742-5468/aa569b, https://doi.org/10.1088/1742-5468/aa569b; arXiv:1610.08320, https://arxiv.org/abs/1610.08320
  2. B. Derrida, M. R. Evans, V. Hakim and V. Pasquier, 'Exact solution of a 1D asymmetric exclusion model using a matrix formulation', J. Phys. A: Math. Gen. 26, 1493 (1993). DOI: 10.1088/0305-4470/26/7/011, https://doi.org/10.1088/0305-4470/26/7/011
  3. B. Derrida, M. R. Evans and K. Mallick, 'Exact diffusion constant of a one-dimensional asymmetric exclusion model with open boundaries', J. Stat. Phys. 79, 833 (1995). DOI: 10.1007/BF02181206, https://doi.org/10.1007/BF02181206
  4. B. Derrida and J. L. Lebowitz, 'Exact Large Deviation Function in the Asymmetric Exclusion Process', Phys. Rev. Lett. 80, 209 (1998). DOI: 10.1103/PhysRevLett.80.209, https://doi.org/10.1103/PhysRevLett.80.209
  5. J. de Gier and F. H. L. Essler, 'Large Deviation Function for the Current in the Open Asymmetric Simple Exclusion Process', Phys. Rev. Lett. 107, 010602 (2011). DOI: 10.1103/PhysRevLett.107.010602, https://doi.org/10.1103/PhysRevLett.107.010602; arXiv:1101.3235, https://arxiv.org/abs/1101.3235
  6. A. Lazarescu and K. Mallick, 'An exact formula for the statistics of the current in the TASEP with open boundaries', J. Phys. A: Math. Theor. 44, 315001 (2011). DOI: 10.1088/1751-8113/44/31/315001, https://doi.org/10.1088/1751-8113/44/31/315001; arXiv:1104.5089, https://arxiv.org/abs/1104.5089
  7. M. Gorissen, A. Lazarescu, K. Mallick and C. Vanderzande, 'Exact Current Statistics of the Asymmetric Simple Exclusion Process with Open Boundaries', Phys. Rev. Lett. 109, 170601 (2012). DOI: 10.1103/PhysRevLett.109.170601, https://doi.org/10.1103/PhysRevLett.109.170601; arXiv:1207.6879, https://arxiv.org/abs/1207.6879
  8. A. Lazarescu, 'Matrix ansatz for the fluctuations of the current in the ASEP with open boundaries', J. Phys. A: Math. Theor. 46, 145003 (2013). DOI: 10.1088/1751-8113/46/14/145003, https://doi.org/10.1088/1751-8113/46/14/145003; arXiv:1212.3366, https://arxiv.org/abs/1212.3366
  9. A. Lazarescu and V. Pasquier, 'Bethe Ansatz and Q-operator for the open ASEP', J. Phys. A: Math. Theor. 47, 295202 (2014). DOI: 10.1088/1751-8113/47/29/295202, https://doi.org/10.1088/1751-8113/47/29/295202; arXiv:1403.6963, https://arxiv.org/abs/1403.6963
  10. A. Lazarescu, 'The Physicist's Companion to Current Fluctuations: One-Dimensional Bulk-Driven Lattice Gases', J. Phys. A: Math. Theor. 48, 503001 (2015). DOI: 10.1088/1751-8113/48/50/503001, https://doi.org/10.1088/1751-8113/48/50/503001; arXiv:1507.04179, https://arxiv.org/abs/1507.04179
  11. N. Reshetikhin, 'Jackson-type integrals, Bethe vectors, and solutions to a difference analog of the Knizhnik-Zamolodchikov system', Lett. Math. Phys. 26, 153 (1992). DOI: 10.1007/BF00420749, https://doi.org/10.1007/BF00420749
  12. N. Reshetikhin and A. Varchenko, 'Quasiclassical asymptotics of solutions to the KZ equations' (1994). arXiv:hep-th/9402126, https://arxiv.org/abs/hep-th/9402126
  13. L. Cantini, 'Asymmetric Simple Exclusion Process with open boundaries and Koornwinder polynomials' (2015). arXiv:1506.00284, https://arxiv.org/abs/1506.00284
  14. L. Cantini, J. de Gier and M. Wheeler, 'Matrix product formula for Macdonald polynomials', J. Phys. A: Math. Theor. 48, 384001 (2015). DOI: 10.1088/1751-8113/48/38/384001, https://doi.org/10.1088/1751-8113/48/38/384001; arXiv:1505.00287, https://arxiv.org/abs/1505.00287
  15. N. Crampe, K. Mallick, E. Ragoucy and M. Vanicat, 'Inhomogeneous discrete-time exclusion processes', J. Phys. A: Math. Theor. 48, 484002 (2015). DOI: 10.1088/1751-8113/48/48/484002, https://doi.org/10.1088/1751-8113/48/48/484002; arXiv:1506.04874, https://arxiv.org/abs/1506.04874
  16. L. Cantini, A. Garbali, J. de Gier and M. Wheeler, 'Koornwinder polynomials and the stationary multi-species asymmetric exclusion process with open boundaries', J. Phys. A: Math. Theor. 49, 444002 (2016). DOI: 10.1088/1751-8113/49/44/444002, https://doi.org/10.1088/1751-8113/49/44/444002; arXiv:1607.00039, https://arxiv.org/abs/1607.00039
  17. M. Vanicat, 'An integrabilist approach of out-of-equilibrium statistical physics models' (PhD thesis, 2017). arXiv:1708.02440, https://arxiv.org/abs/1708.02440
  18. C. Finn, E. Ragoucy and M. Vanicat, 'Matrix product solution to multi-species ASEP with open boundaries', J. Stat. Mech. 2018, 043201 (2018). DOI: 10.1088/1742-5468/aab1b5, https://doi.org/10.1088/1742-5468/aab1b5; arXiv:1712.06809, https://arxiv.org/abs/1712.06809
  19. N. Crampe and R. I. Nepomechie, 'Equivalent T-Q relations and exact results for the open TASEP', J. Stat. Mech. 2018, 103105 (2018). DOI: 10.1088/1742-5468/aae2e0, https://doi.org/10.1088/1742-5468/aae2e0; arXiv:1806.07748, https://arxiv.org/abs/1806.07748
  20. B. Buča, J. P. Garrahan, T. Prosen and M. Vanicat, 'Exact large deviation statistics and trajectory phase transition of a deterministic boundary driven cellular automaton', Phys. Rev. E 100, 020103 (2019). DOI: 10.1103/PhysRevE.100.020103, https://doi.org/10.1103/PhysRevE.100.020103; arXiv:1901.00845, https://arxiv.org/abs/1901.00845
  21. R. Brak and W. Moore, 'Bi-orthogonal Polynomials and the Five parameter Asymmetric Simple Exclusion Process' (2019). arXiv:1902.06373, https://arxiv.org/abs/1902.06373
  22. X. Zhang, F.-K. Wen and J. de Gier, 'T-Q relations for the integrable two-species asymmetric simple exclusion process with open boundaries', J. Stat. Mech. 2019, 014001 (2019). DOI: 10.1088/1742-5468/aaeb4a, https://doi.org/10.1088/1742-5468/aaeb4a; arXiv:1808.05880, https://arxiv.org/abs/1808.05880
  23. M. Vanicat, E. Bertin, V. Lecomte and E. Ragoucy, 'Mapping current and activity fluctuations in exclusion processes: consequences and open questions', SciPost Phys. 10, 028 (2021). DOI: 10.21468/SciPostPhys.10.2.028, https://doi.org/10.21468/SciPostPhys.10.2.028; arXiv:2011.02202, https://arxiv.org/abs/2011.02202
  24. A. Garbali, J. de Gier, W. Mead and 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, https://doi.org/10.1007/s00023-024-01484-5; arXiv:2312.14348, https://arxiv.org/abs/2312.14348
  25. A. Garbali, W. Guo and M. Wheeler, 'Rational symmetric functions from the Izergin-Korepin 19-vertex model' (2024). arXiv:2412.18085, https://arxiv.org/abs/2412.18085
  26. S. Corteel, O. Mandelshtam and L. Williams, 'Rhombic staircase tableaux and Koornwinder polynomials', Math. Z. 308, 45 (2024). DOI: 10.1007/s00209-024-03596-4, https://doi.org/10.1007/s00209-024-03596-4; arXiv:2312.17469, https://arxiv.org/abs/2312.17469
  27. D. Wu, 'Stable Limit DAHA of type (C∨,C)(C^{\vee},C) and Stable Limit Koornwinder Polynomials' (2026). arXiv:2607.12276, https://arxiv.org/abs/2607.12276
  28. J. He and M. Wheeler, 'Free boundary q-Whittaker and Hall-Littlewood processes' (2025). arXiv:2512.02267, https://arxiv.org/abs/2512.02267