Pith. sign in

REVIEW 3 major objections 3 minor 1 cited by

Efficient matrix-product-state preparation of highly entangled trial states: Weak Mott insulators on the triangular lattice revisited

T0 review · 3 major / 3 minor · reviewed 2026-08-27 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The triangular-lattice J-K ring-exchange model likely lacks a gapless U(1) spinon Fermi surface on 4- and 6-leg cylinders; DMRG warm-started from a Gutzwiller-projected trial state finds gapped, flat-entanglement ground states.

desk verdict The Gutzwiller zipper is a genuine technical advance and the 4-leg result is well-supported, but the 6-leg conclusion is undercut by an unconverged trial state and missing bond-dimension scaling. read the letter →

arxiv 2009.12435 v2 pith:TJHZ2P2E submitted 2020-09-25 cond-mat.str-el quant-ph

Amir M Aghaei , Bela Bauer , Kirill Shtengel , Ryan V. Mishmash This is my paper · ORCID
classification cond-mat.str-elquant-ph
keywords matrixproductstatesdensityrenormalizationgroupGutzwillerprojectionspinonFermisurfacespinBosemetaltriangularlatticering-exchangemodelentanglemententropy
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Some quantum spin liquids, the spin Bose metals, host fractionalized 'spinon' particles that form a gapless Fermi surface, and they are the leading candidate for the insulating ground state of weakly Mott insulating triangular-lattice magnets. Their hallmark is logarithmic growth of entanglement, which makes them almost impossible to study with standard DMRG because the required matrix-product-state bond dimension grows exponentially with system width. This paper develops an efficient 'Gutzwiller zipper' for building an MPS representation of the Gutzwiller-projected Fermi surface (GPFS) trial wave function, and uses that highly entangled trial state to 'warm start' DMRG; on the 2-leg ladder, where the spin Bose metal is well established, the method preserves the expected central charge $c=3$ and converges far faster than random initialization. On 4- and 6-leg cylinders of the J-K spin model (Heisenberg exchange plus four-site ring exchange) at $K/J=0.6$, the same procedure makes the entanglement entropy collapse within a few sweeps to flat or nearly flat scaling, which the authors read as the true ground state being gapped and the previously proposed $c=5$ and $c=9$ spin Bose metal phases being unstable. If correct, the result contradicts earlier DMRG claims about a central model of weak Mott insulators and gives a general tool for testing gapless spin liquid proposals.

What carries the argument

The load-bearing object is the 'Gutzwiller zipper': an algorithm that combines two fermionic Gaussian MPSs (one for each spinon flavor) with the Gutzwiller projection, which removes empty and doubly occupied sites, and truncates the result to a target bond dimension $M$ in a single left-to-right sweep, at cost $O(N m^2 M^2)$. It is paired with a warm-start strategy: initialize DMRG with this highly entangled trial MPS. The logic is that DMRG energy optimization will preserve the universal entanglement scaling (central charge $c=2N_{\mathrm{slices}}-1$) if the true ground state is in the trial state's universality class, but will rapidly shed entanglement if the true ground state is a less-entangled instability. The quantitative argument is carried by fits of the von Neumann entropy to the finite-size conformal formula $S_1(\ell,N)=\frac{c}{6}\log\big(\frac{N}{\pi}\sin\frac{\pi\ell}{N}\big)+A'$.

What would settle it

Converge the 6-leg DMRG calculation with a fully converged GPFS trial state (bond dimensions $m\geq 2000$ and $M\geq 10000$) on longer cylinders and check whether the entanglement entropy still collapses to a flat curve; if the converged entanglement scaling instead approaches $c=9$ (or $c=8$), the paper's central claim would be falsified. A complementary check is to measure correlation functions in the 4-leg DMRG ground state: the paper's claim predicts exponentially decaying spin correlations and a finite gap, whereas the spin Bose metal predicts power-law correlations and gapless modes.

Watch

Extended reading notes

Core claim

The paper's central claim is that the simplest isotropic triangular-lattice J-K spin model with four-site ring exchange does not host a fully gapless U(1) spinon Fermi surface (spin Bose metal) ground state on 4- and 6-leg wide ladders, contrary to previous DMRG and variational studies. The evidence is the behavior of the entanglement entropy when DMRG is initialized from a faithful GPFS MPS: if the true ground state were in the same universality class, energy optimization should leave the trial state's $c=2N_{\mathrm{slices}}-1$ scaling intact, whereas on the 4-leg cylinder the entropy collapses within two sweeps to a flat curve consistent with $c=0$, and on the 6-leg cylinder it drops to a nearly flat curve with strong rung-to-rung oscillations. The authors take this entropy shedding to mean that the DMRG ground state is less entangled than the trial state, i.e., gapped, and they argue that an eventual $c=9$ (or $c=8$ for a Z$_2$ spinon Fermi surface) outcome on 6 legs is unlikely, noting explicitly that the 6-leg GPFS trial state itself is not fully converged.

Load-bearing premise

The argument rests on the assumption that initializing DMRG with a highly entangled trial state preserves that entanglement when the true ground state is in the same universality class and sheds it when the true ground state is less entangled; this behavior was verified on the 2-leg ladder but is assumed on the 4- and 6-leg cylinders, and the 6-leg conclusion also assumes that the not-fully-converged GPFS MPS still represents the spin Bose metal universality class well enough for the observed drop in entanglement to be meaningful.

Editorial extensions

If this is right

  • At $K/J=0.6$ on the 4-leg cylinder, the DMRG ground state of the J-K model is likely fully gapped, with flat entanglement scaling, rather than the $c=5$ spin Bose metal proposed in earlier work.
  • On the 6-leg cylinder, an eventual $c=9$ spinon Fermi surface (or $c=8$ Z$_2$ spinon Fermi surface) is unlikely; the ground state appears to have lower, nearly flat entanglement.
  • The Gutzwiller-zipper MPS representation of the GPFS trial state reproduces the spin structure factor obtained by variational Monte Carlo, so the approximate trial state faithfully captures long-distance correlations even when its entanglement is not fully converged.
  • The warm-start strategy reproduces the full 2-leg $J_1$-$J_2$-$K$ phase diagram after only a couple of DMRG sweeps, including phases whose central charge is lower than that of the seed trial state.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial inference: If the 4- and 6-leg results survive larger systems, the Gutzwiller-projected Fermi surface may not be the correct starting point for weak Mott insulators on the triangular lattice, and the natural alternatives are gapped chiral or Z$_2$ spin liquid states.
  • Editorial inference: The same entanglement-shedding diagnostic could be used as a general instability test for any proposed gapless phase: initialize DMRG with the most entangled plausible trial state and watch whether O(1) sweeps destroy its universal entanglement.
  • Editorial inference: Because the zipper produces an explicit MPS for a projected wave function, it makes R\'enyi entropies $S_\alpha$ at arbitrary $\alpha$ computable for such trial states, which may sharpen distinctions between gapless and topologically ordered phases.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Request a human review

A listed scientist reviews the paper for a fee and the review publishes here regardless of verdict. See the reviewers or get listed.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 3 minor

Summary. The manuscript introduces a 'Gutzwiller zipper' algorithm that converts Gutzwiller-projected Fermi surface (GPFS) trial wave functions into matrix product states with bond dimension M, and uses these as initial states ('warm start') for DMRG simulations of the triangular-lattice J-K ring-exchange model. Benchmarking on the 2-leg ladder, the authors reproduce the previously established c=3 spin Bose metal and show that DMRG initialized with the GPFS MPS converges much faster than with random initialization. On 4-leg cylinders, the GPFS MPS is converged to c≈5 and the DMRG ground state at K/J=0.6 exhibits nearly flat entanglement entropy, which the authors interpret as evidence that the true ground state is gapped rather than the c=5 U(1) SFS proposed by Block et al. On 6-leg cylinders, the GPFS MPS cannot be fully converged to the expected c=9, and DMRG again produces a nearly flat entropy profile, leading the authors to argue that a c=9 SFS is unlikely. The paper concludes that the simplest J-K model likely does not host a gapless U(1) spinon Fermi surface on 4- and 6-leg ladders and presents the state-preparation strategy as a general tool.

Significance. If correct, the paper overturns earlier DMRG-based claims of a U(1) spinon Fermi surface in the J-K model on 4- and 6-leg cylinders and provides a new numerical route for studying highly entangled trial states. The Gutzwiller zipper is a practical algorithmic advance: it produces an MPS representation of the GPFS state with cost scaling O(N M^2 m^2), and the agreement of the MPS structure factor with direct variational Monte Carlo (Fig. 12) is a valuable cross-check. The 2-leg benchmark (Fig. 9) is a well-designed validation of the warm-start heuristic across several phases. The 4-leg result is supported by convergence of the trial state to c≈5, near-flat DMRG entropy, agreement with random-initialization DMRG, and a structure factor consistent with previous work. The 6-leg conclusion, by contrast, is provisional because both the trial and final states are admitted to be unconverged in bond dimension.

major comments (3)
  1. [Fate of the SFS in the triangular lattice J-K model (Fig. 4)] The 4-leg conclusion would be considerably strengthened by a convergence analysis of S1(N/2) for the final DMRG state as a function of bond dimension M. The data in Fig. 4 and footnote 51 establish invariance under further sweeps and consistency with random initialization, but a gapless c=5 state optimized at fixed M can develop a flat S1 profile because of the finite-M entanglement cap; the demonstration that the GPFS MPS reaches c≈5 at the same M is suggestive but does not by itself prove that the DMRG-optimized state has converged to a genuinely gapped state. In light of Eq. (3), a plot of S1(N/2) versus M (e.g., M=1000, 2000, 3000, 4000) for Lx=24 and 42 would control for this artifact.
  2. [Fate of the SFS in the triangular lattice J-K model, 6-leg ladder (Fig. 5) and Appendix C.3] The 6-leg conclusion rests on data that the manuscript itself labels as not fully converged: the GPFS MPS has not reached the expected c=9 (main text; Fig. 11), and the final DMRG entropy 'is still not fully converged in M' (Appendix C.3). Because the trial state is not a faithful MPS representation of the SFS universality class, the warm-start argument does not allow a clean interpretation of the subsequent entanglement drop: the drop could reflect DMRG correcting an incompletely converged trial state rather than an instability of the U(1) SFS. The paper's cautious wording ('seem unlikely') is appropriate, but the abstract's joint claim for four- and six-leg ladders is stronger than the 6-leg evidence supports. Converging the 6-leg GPFS to c=9, or providing an M-scaling analysis of the final DMRG S1(N/2) showing a plateau well below the c=9 expectation, is needed.
  3. [State preparation strategy] The central inference--that a drop in entanglement after warm-start DMRG implies that the true ground state has lower central charge--relies on an assumption that is validated only on the 2-leg ladder (Fig. 9). The benchmark does not cover the regime most relevant to this paper's claims, namely a gapless ground state on 4- or 6-leg cylinders at finite M, where DMRG can under-resolve the entanglement of a gapless state. An explicit test of the heuristic on a 4-leg system whose ground state is independently known to be gapless (if such a point exists in the J-K phase diagram or a related model) would substantially increase confidence in the 4- and 6-leg conclusions.
minor comments (3)
  1. [Appendix C.1 (Fig. 9 caption)] The word 'reminiscient' should be 'reminiscent' in the caption text.
  2. [References] Reference [66] is missing a closing bracket: it reads 'arXiv:2009.04129 [cond-mat (2020)' and should read 'arXiv:2009.04129 [cond-mat] (2020)'.
  3. [Introduction] The notation 'M∼eS∼(ALαx)Ly' is ambiguous; suggest writing 'S∼L_y log L_x' and 'M∼exp(S)∼(A L_x^α)^{L_y}' with a clear definition of the constants.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the GPFS warm-start result is an independent numerical outcome, not an input refit.

full rationale

The central claim—that DMRG on 4- and 6-leg J-K cylinders yields flat entanglement rather than c=5/9 SFS scaling—does not reduce to its inputs. The GPFS trial state is constructed from a free-fermion parton Hamiltonian (Eq. B2) with hand-chosen hopping, then Gutzwiller projected via the zipper; its entanglement is measured (Fig. 3) and serves as a high-entanglement initial condition. The final DMRG entanglement (Figs. 4-5) is a separately computed quantity, not obtained by fitting the trial state's c. The c=0 conclusion is cross-checked by random initialization on 4x24 (footnote 51) and by VMC agreement of structure factors (Fig. 12). Self-citations to Refs. [14,20,21,25,29,40] are either prior claims being tested (the paper's result contradicts [21,29]) or method provenance (Ref. [40] for the zipper lineage); none is a load-bearing uniqueness theorem or a fitted parameter renamed as prediction. The admitted non-convergence of the 6-leg GPFS MPS (m=1200, M=6000) and of the final DMRG entropy in M (App. C.3, Fig. 11) are numerical convergence limitations that could affect correctness, but they do not make the derivation circular. No step in the paper defines the predicted quantity in terms of the observed quantity, and no uniqueness result from the authors is invoked to forbid alternatives.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The central physics claim introduces no new free parameters: trial states are built from standard free-fermion mean-field ansatze with hand-chosen hopping, and Hamiltonian couplings come from prior work. The main numerical parameters are bond dimensions and central-charge fit windows. No new particles, forces, or other entities are postulated.

free parameters (3)
  • Parton hopping ratio t2/t1 = 0.7
    Hand-picked in the 2-leg benchmark to produce a generic two-band spinon Fermi surface; not used in the 4- and 6-leg calculations, so it does not directly affect the central negative claim.
  • MPS bond dimensions m and M = m=200-1200, M=900-6000 across simulations
    Computational truncation parameters chosen for convergence; the 6-leg GPFS trial state (m=1200, M=6000) is explicitly not fully converged, which weakens the 6-leg conclusion.
  • Calabrese-Cardy fit exclusion windows = 8 sites/side on 2-leg; 7 rungs on 4-leg; 2 rungs on PBC
    Post hoc exclusion of boundary data in central charge fits; the paper notes the choice significantly impacts the fitted c (Appendix B.2).
assumptions (5)
  • domain assumption The Calabrese-Cardy formula S1(l,N) = (c/6) log((N/pi) sin(pi l/N)) + A' describes entanglement entropy of gapless states on these open cylinders.
    Invoked in Eq. (B5) to extract all central charges from entanglement data.
  • domain assumption The Gutzwiller-projected Fermi surface wave function faithfully represents the universal physics of the U(1) spinon Fermi surface.
    Identifies the GPFS MPS as the correct highly entangled mother state; standard in the spin liquid literature (Refs. [19,22]).
  • ad hoc to paper DMRG energy optimization preserves the entanglement scaling of a highly entangled trial state when the true ground state is in the same universality class; a drop implies a lower-entanglement ground state.
    The central premise of the warm-start strategy; validated on the 2-leg ladder but assumed on 4/6-leg.
  • standard math The Gaussian MPS construction and SVD truncation in the Gutzwiller zipper produce a controlled approximation to the exact GPFS state.
    Underlies all trial states; Appendix A.4 acknowledges that the truncation is not globally optimal.
  • domain assumption The J-K ring-exchange Hamiltonian is the appropriate minimal model for the weak Mott insulator regime on the triangular lattice.
    Adopted from Refs. [19,21,27-31]; the paper does not re-derive the model.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Efficient matrix-product-state preparation of highly entangled trial states: Weak Mott insulators on the triangular lattice revisited." pith.science (2026). https://pith.science/paper/TJHZ2P2E

@misc{pith2026200912435,
  author       = {Pith},
  title        = {Pith review of: Efficient matrix-product-state preparation of highly entangled trial states: Weak Mott insulators on the triangular lattice revisited},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/TJHZ2P2E}},
  note         = {Machine review of arXiv:2009.12435}
}
abstract

Using tensor network states to unravel the physics of quantum spin liquids in minimal, yet generic microscopic spin or electronic models remains notoriously challenging. A prominent open question concerns the nature of the insulating ground state of two-dimensional half-filled Hubbard-type models on the triangular lattice in the vicinity of the Mott metal-insulator transition, a regime which can be approximated microscopically by a spin-1/2 Heisenberg model supplemented with additional "ring-exchange" interactions. Using a novel and efficient state preparation technique whereby we initialize full density matrix renormalization group (DMRG) calculations with highly entangled Gutzwiller-projected Fermi surface trial wave functions, we show -- contrary to previous works -- that the simplest triangular lattice $J$-$K$ spin model with four-site ring exchange likely does not harbor a fully gapless U(1) spinon Fermi surface (spin Bose metal) phase on four- and six-leg wide ladders. Our methodology paves the way to fully resolve with DMRG other controversial problems in the fields of frustrated quantum magnetism and strongly correlated electrons.

Figures

Figures reproduced from arXiv: 2009.12435 by the authors.

Figure 1
Figure 1. FIG. 1. Illustration of a single step of the [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2. Entanglement entropy [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 5
Figure 5. FIG. 5. Entanglement entropy scaling of the GPFS MPS (which is [PITH_FULL_IMAGE:figures/full_fig_p004_5.png] view at source ↗
Figures from the paper (7 more)
Figure 4
Figure 4. Figure 4: FIG. 4. Entanglement entropy scaling (top) and spin structure factor [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]
Figure 6
Figure 6. Figure 6: FIG. 6. Entanglement entropy of the GPFS state obtained from a [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]
Figure 7
Figure 7. Figure 7: FIG. 7. The type of triangular lattice clusters we consider, here drawn [PITH_FULL_IMAGE:figures/full_fig_p009_7.png]
Figure 8
Figure 8. Figure 8: FIG. 8. The Fermi sea of the 4-leg (left) and 6-leg (right) triangular [PITH_FULL_IMAGE:figures/full_fig_p009_8.png]
Figure 9
Figure 9. Figure 9: FIG. 9. Exploring the phase diagram of the 2-leg ladder [PITH_FULL_IMAGE:figures/full_fig_p010_9.png]
Figure 10
Figure 10. Figure 10: FIG. 10. Entanglement entropy [PITH_FULL_IMAGE:figures/full_fig_p011_10.png]
Figure 11
Figure 11. Figure 11: FIG. 11. The von Neumann entanglement entropy as a function of [PITH_FULL_IMAGE:figures/full_fig_p012_11.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Dirac and chiral spin liquids on spin-1/2 square-lattice Heisenberg antiferromagnet

    cond-mat.str-el 2024-11 conditional novelty 6.0 of 10

    The spin-1/2 square-lattice J1-J2 Heisenberg antiferromagnet at J2=0.5J1 is identified, via DMRG and parton wave-function fidelity, as a gapless Z2 Dirac quantum spin liquid.

Reference graph

Works this paper leans on

98 extracted references · 62 canonical work pages · cited by 1 Pith paper

  1. [40]

    Bauer, B

    B. Bauer, B. P. Keller, S. Trebst, and A. W. W. Ludwig, Phys. Rev. B 99, 035155 (2019)

    work page 2019

  2. [1]

    Balents, Nature 464, 199 (2010)

    L. Balents, Nature 464, 199 (2010)

    2010

  3. [2]

    Savary and L

    L. Savary and L. Balents, Rep. Prog. Phys. 80, 016502 (2017)

    2017

  4. [3]

    Y . Zhou, K. Kanoda, and T.-K. Ng, Rev. Mod. Phys.89, 025003 (2017)

    2017

  5. [4]

    Broholm, R

    C. Broholm, R. J. Cava, S. A. Kivelson, D. G. Nocera, M. R. Norman, and T. Senthil, Science 367, eaay0668 (2020)

    2020

  6. [5]

    P. W. Anderson, Mater. Res. Bull. 8, 153 (1973)

    1973

  7. [6]

    P. W. Anderson, Science 235, 1196 (1987)

    1987

  8. [7]

    Or ´us, Ann

    R. Or ´us, Ann. Phys. 349, 117 (2014)

    2014

Show all 98 references
  1. [8]

    Cincio and G

    L. Cincio and G. Vidal, Phys. Rev. Lett. 110, 067208 (2013)

    2013

  2. [9]

    Poilblanc, J

    D. Poilblanc, J. I. Cirac, and N. Schuch, Phys. Rev. B 91, 224431 (2015)

    2015

  3. [10]

    Y .-C. He, D. N. Sheng, and Y . Chen, Phys. Rev. B89, 075110 (2014)

    2014

  4. [11]

    Misguich, C

    G. Misguich, C. Lhuillier, B. Bernu, and C. Waldtmann, Phys. Rev. B 60, 1064 (1999)

    1999

  5. [12]

    LiMing, G

    W. LiMing, G. Misguich, P. Sindzingre, and C. Lhuillier, Phys. Rev. B 62, 6372 (2000)

    2000

  6. [13]

    Zhang, T

    Y . Zhang, T. Grover, and A. Vishwanath, Phys. Rev. Lett.107, 067202 (2011)

    2011

  7. [14]

    R. V . Mishmash and O. I. Motrunich, Phys. Rev. B 94, 081110 (2016)

    2016

  8. [15]

    W.-J. Hu, Y . Zhang, A. H. Nevidomskyy, E. Dagotto, Q. Si, and H.-H. Lai, Phys. Rev. Lett. 124, 237201 (2020)

    2020

  9. [16]

    Stoudenmire and S

    E. Stoudenmire and S. R. White, Annu. Rev. Condens. Matter Phys. 3, 111 (2012)

    2012

  10. [17]

    In the case of a gapped spin liquid, one would expect α = 0

  11. [18]

    Lee and P

    S.-S. Lee and P. A. Lee, Phys. Rev. Lett. 95, 036403 (2005)

    2005

  12. [19]

    O. I. Motrunich, Phys. Rev. B 72, 045105 (2005)

    2005

  13. [20]

    D. N. Sheng, O. I. Motrunich, and M. P. A. Fisher, Phys. Rev. B 79, 205112 (2009)

    2009

  14. [21]

    M. S. Block, D. N. Sheng, O. I. Motrunich, and M. P. A. Fisher, Phys. Rev. Lett. 106, 157202 (2011)

    2011

  15. [22]

    For some potential caveats, see Ref. [67]

  16. [23]

    Senthil, Phys

    T. Senthil, Phys. Rev. B 78, 045109 (2008)

    2008

  17. [24]

    Lai and O

    H.-H. Lai and O. I. Motrunich, Phys. Rev. B81, 045105 (2010)

    2010

  18. [25]

    R. V . Mishmash, I. Gonz ´alez, R. G. Melko, O. I. Motrunich, and M. P. A. Fisher, Phys. Rev. B91, 235140 (2015)

    2015

  19. [26]

    Kanoda and R

    K. Kanoda and R. Kato, Annu. Rev. Condens. Matter Phys. 2, 167 (2011)

    2011

  20. [27]

    Grover, N

    T. Grover, N. Trivedi, T. Senthil, and P. A. Lee, Phys. Rev. B 81, 245121 (2010)

    2010

  21. [28]

    R. V . Mishmash, J. R. Garrison, S. Bieri, and C. Xu, Phys. Rev. Lett. 111, 157203 (2013)

    2013

  22. [29]

    W.-Y . He, X. Y . Xu, G. Chen, K. Law, and P. A. Lee, Phys. Rev. Lett. 121, 046401 (2018)

    2018

  23. [30]

    A. H. MacDonald, S. M. Girvin, and D. Yoshioka, Phys. Rev. B 37, 9753 (1988)

    1988

  24. [31]

    H.-Y . Yang, A. M. L¨auchli, F. Mila, and K. P. Schmidt, Phys. Rev. Lett. 105, 267204 (2010)

    2010

  25. [32]

    Calabrese and J

    P. Calabrese and J. Cardy, J. Stat. Mech.: Theory Exp. 6, 06002 (2004)

    2004

  26. [33]

    S. D. Geraedts, M. P. Zaletel, R. S. K. Mong, M. A. Metlitski, A. Vishwanath, and O. I. Motrunich, Science 352, 197 (2016)

    2016

  27. [34]

    Gondzio, Math

    J. Gondzio, Math. Program. 83, 125 (1998)

    1998

  28. [35]

    J. R. McClean, S. Boixo, V . N. Smelyanskiy, R. Babbush, and H. Neven, Nat. Commun. 9, 4812 (2018)

    2018

  29. [36]

    Zhou, S.-T

    L. Zhou, S.-T. Wang, S. Choi, H. Pichler, and M. D. Lukin, Phys. Rev. X 10, 021067 (2020)

    2020

  30. [37]

    D. J. Egger, J. Marecek, and S. Woerner, arXiv:2009.10095 [quant-ph] (2020)

    2020 arXiv

  31. [38]

    Reiher, N

    M. Reiher, N. Wiebe, K. M. Svore, D. Wecker, and M. Troyer, Proc. Natl. Acad. Sci. U.S.A. 114, 7555 (2017)

    2017

  32. [39]

    N. M. Tubman, C. Mejuto-Zaera, J. M. Epstein, D. Hait, D. S. Levine, W. Huggins, Z. Jiang, J. R. McClean, R. Babbush, M. Head-Gordon, and K. B. Whaley, arXiv:1809.05523 [cond- mat] (2018)

    2018 arXiv

  33. [41]

    The U(1) SFS can be thought of as a “mother” state [68]; fur- thermore, it is the most highly entangled plausible universality class of the model

  34. [42]

    Although some counterexamples exist; see, e.g., Ref. [40]

  35. [43]

    M. T. Fishman and S. R. White, Phys. Rev. B 92, 075132 (2015)

    2015

  36. [44]

    For the SU(2) invariant U(1) SFS ansatz, the mean-field hop- ping parameters for the two species of spinons are identical

  37. [45]

    Wecker, M

    D. Wecker, M. B. Hastings, N. Wiebe, B. K. Clark, C. Nayak, and M. Troyer, Phys. Rev. A92, 062318 (2015)

    2015

  38. [46]

    I. D. Kivlichan, J. McClean, N. Wiebe, C. Gidney, A. Aspuru- Guzik, G. K.-L. Chan, and R. Babbush, Phys. Rev. Lett. 120, 110501 (2018)

    2018

  39. [47]

    Arute et al., Science 369, 1084 (2020)

    F. Arute et al., Science 369, 1084 (2020)

    2020

  40. [48]

    For more details of our approach and simulations plus addi- tional supporting data, please see the Appendices

  41. [49]

    Y .-H. Wu, L. Wang, and H.-H. Tu, Phys. Rev. Lett.124, 246401 (2020)

    2020

  42. [50]

    Jin, H.-H

    H.-K. Jin, H.-H. Tu, and Y . Zhou, Phys. Rev. B 101, 165135 (2020)

    2020

  43. [51]

    For N = 4× 24, we have also checked that random state initialization gives the same finalS1 values after many sweeps

    The data is basically invariant under further sweeps. For N = 4× 24, we have also checked that random state initialization gives the same finalS1 values after many sweeps

  44. [52]

    Gros, Ann

    C. Gros, Ann. Phys. 189, 53 (1989)

    1989

  45. [53]

    VIII of the Supplemental Material of Ref

    Note, however, that the models are not exactly the same; see discussion in Sec. VIII of the Supplemental Material of Ref. [29]

  46. [54]

    Petrica, B.-X

    G. Petrica, B.-X. Zheng, G. K.-L. Chan, and B. K. Clark, arXiv:2009.00064 [cond-mat] (2020)

    2020 arXiv

  47. [55]

    Szasz, J

    A. Szasz, J. Motruk, M. P. Zaletel, and J. E. Moore, Phys. Rev. X 10, 021042 (2020)

    2020

  48. [56]

    Shirakawa, T

    T. Shirakawa, T. Tohyama, J. Kokalj, S. Sota, and S. Yunoki, Phys. Rev. B 96, 205130 (2017)

    2017

  49. [57]

    X. Y . Xu, K. Law, and P. A. Lee, Phys. Rev. Lett.122, 167001 (2019)

    2019

  50. [58]

    [40, 69–82]

    See, e.g., Refs. [40, 69–82]

  51. [59]

    S. Yan, D. A. Huse, and S. R. White, Science 332, 1173 (2011). 6

    2011

  52. [60]

    Kolley, S

    F. Kolley, S. Depenbrock, I. P. McCulloch, U. Schollw¨ock, and V . Alba, Phys. Rev. B91, 104418 (2015)

    2015

  53. [61]

    Y .-C. He, M. P. Zaletel, M. Oshikawa, and F. Pollmann, Phys. Rev. X 7, 031020 (2017)

    2017

  54. [62]

    Hu, S.-S

    W.-J. Hu, S.-S. Gong, W. Zhu, and D. N. Sheng, Phys. Rev. B 92, 140403 (2015)

    2015

  55. [63]

    Zhu and S

    Z. Zhu and S. R. White, Phys. Rev. B 92, 041105 (2015)

    2015

  56. [64]

    S. Hu, W. Zhu, S. Eggert, and Y .-C. He, Phys. Rev. Lett. 123, 207203 (2019)

    2019

  57. [65]

    Baiardo and M

    A. Baiardo and M. Reiher, arXiv:2009.02614 [cond-mat] (2020)

    2020 arXiv

  58. [66]

    Jin, H.-H

    H.-K. Jin, H.-H. Tu, and Y . Zhou, arXiv:2009.04129 [cond-mat (2020)

    2020 arXiv

  59. [67]

    Tay and O

    T. Tay and O. I. Motrunich, Phys. Rev. B 83, 235122 (2011)

    2011

  60. [68]

    Hermele, T

    M. Hermele, T. Senthil, and M. P. A. Fisher, Phys. Rev. B 72, 104404 (2005)

    2005

  61. [69]

    O. I. Motrunich and M. P. A. Fisher, Phys. Rev. B 75, 235116 (2007)

    2007

  62. [70]

    D. N. Sheng, O. I. Motrunich, S. Trebst, E. Gull, and M. P. A. Fisher, Phys. Rev. B78, 054520 (2008)

    2008

  63. [71]

    M. S. Block, R. V . Mishmash, R. K. Kaul, D. N. Sheng, O. I. Motrunich, and M. P. A. Fisher, Phys. Rev. Lett. 106, 046402 (2011)

    2011

  64. [72]

    R. V . Mishmash, M. S. Block, R. K. Kaul, D. N. Sheng, O. I. Motrunich, and M. P. A. Fisher, Phys. Rev. B 84, 245127 (2011)

    2011

  65. [73]

    Jiang, M

    H.-C. Jiang, M. S. Block, R. V . Mishmash, J. R. Garrison, D. N. Sheng, O. I. Motrunich, and M. P. A. Fisher, Nature 493, 39 (2012)

    2012

  66. [74]

    Bieri, L

    S. Bieri, L. Messio, B. Bernu, and C. Lhuillier, Phys. Rev. B 92, 060407 (2015)

    2015

  67. [75]

    S.-S. Gong, W. Zheng, M. Lee, Y .-M. Lu, and D. N. Sheng, Phys. Rev. B 100, 241111 (2019)

    2019

  68. [76]

    Li, Y .-M

    Y .-D. Li, Y .-M. Lu, and G. Chen, Phys. Rev. B 96, 054445 (2017)

    2017

  69. [77]

    M-Aghaei, B

    A. M-Aghaei, B. Bauer, K. Shtengel, and R. V . Mishmash, Phys. Rev. B 98, 054430 (2018)

    2018

  70. [78]

    R. G. Pereira and S. Bieri, SciPost Phys. 4, 004 (2018)

    2018

  71. [79]

    Hickey and S

    C. Hickey and S. Trebst, Nat. Commun. 10, 530 (2019)

    2019

  72. [80]

    N. D. Patel and N. Trivedi, Proc. Natl. Acad. Sci. U.S.A. 116, 12199 (2019)

    2019

  73. [81]

    Jiang, T

    Y .-F. Jiang, T. P. Devereaux, and H.-C. Jiang, Phys. Rev. B100, 165123 (2019)

    2019

  74. [82]

    Keselman, B

    A. Keselman, B. Bauer, C. Xu, and C.-M. Jian, Phys. Rev. Lett. 125, 117202 (2020)

    2020

  75. [83]

    Vidal, Phys

    G. Vidal, Phys. Rev. Lett. 93, 040502 (2004)

    2004

  76. [84]

    Hyatt, J

    K. Hyatt, J. R. Garrison, and B. Bauer, Phys. Rev. Lett. 119, 140502 (2017)

    2017

  77. [85]

    Corboz, G

    P. Corboz, G. Evenbly, F. Verstraete, and G. Vidal, Phys. Rev. A 81, 010303 (2010)

    2010

  78. [86]

    Corboz, R

    P. Corboz, R. Or ´us, B. Bauer, and G. Vidal, Phys. Rev. B 81, 165104 (2010)

    2010

  79. [87]

    Gutzwiller zipper

    P. Virtanen et al., Nature Methods 17, 261 (2020). 7 SUPPLEMENTAL MATERIAL Appendix A: Detailed discussion of the Gutzwiller zipper method In this appendix, we will discuss some details of the “Gutzwiller zipper” approach introduced in the main text. The GPFS model wave functi...

    2020

  80. [88]

    Different ways to generate such a circuit have been studied in the context of state preparation in the quantum computing [45, 46] and tensor network [43, 84] communities

    Parton MPS construction The individual parton MPSs for the↑ and↓ spinons are con- structed by finding a unitary circuit that creates the Slater de- terminant in the local basis from an initial product state MPS; application of the circuit is achieved through standard time evolu...

  81. [89]

    swap tensor

    Fermion sign and symmetries As briefly alluded to in the main text, care must be taken to treat the fermionic sign correctly. To illustrate this issue, we can write the MPS for the↑ partons more explicitly as |ψ↑⟩ = ∑ ⃗ n↑ An↑1 1 ...A n↑N N (c† ↑1)n↑1... (c† ↑N)n↑N|Ω⟩, (A3) whe...

  82. [90]

    Scaling of the Gutzwiller zipper The method starts with the two MPSs in canonical form with the orthogonality center at the first site and bond dimen- sionsm𝓁 at bond𝓁, and the final output is an approximation to the Gutzwiller projected MPS with bond dimensions M𝓁, see Fig. 1. ...

  83. [91]

    Like- wise, the tensorsA𝓁+2 throughAN define a basis for the sites 𝓁 + 2 through N

    Accuracy limitations of the Gutzwiller zipper We can think of the tensorsA1 throughA𝓁−1 of an MPS as defining an (incomplete) basis for the sites 1 through 𝓁− 1; this basis is enumerated by the right index of A𝓁−1; when the tensors are canonical, this basis is orthonormal. Like...

  84. [92]

    Triangular lattice clusters and J-K ring-exchange model The family ofLy-leg triangular lattice clusters that we con- sider is depicted in Fig. 7. On these lattice clusters, we simulate the SU(2) invariant Heisenberg antiferromagnet aug- mented by the four-site cyclic ring-exch...

  85. [93]

    rung cuts

    Entanglement entropy: Definitions and fitting Given the reduced density matrixρA for some subset of the system, the Renyi entanglement entropy is given by Sα(ρA) = 1 1−α log (Trρα A), (B3) whereα is the Renyi index. For α = 1, the conventional von Neumann entanglement entropy is...

  86. [94]

    Spin structure factor The spin structure factor we compute is defined as ⟨Sq· S−q⟩ = 1 N ∑ r,r′ e−iq·(r−r′)⟨Sr· Sr′⟩. (B6) Although we mainly work on open cylinders, we still use this form for the structure factor as the averaging over different “origins” serves to effectively ...

  87. [95]

    warm starting

    Benchmarking “warm starting” DMRG with GPFS MPS on the 2-leg triangular strip ring model In this section, we use theJ1-J2-K ring model on the 2-leg triangular strip [20] as a testbed to benchmark the GPFS state preparation strategy used throughout. We choose a generic 2-band G...

  88. [96]

    GPFS MPS with fully periodic boundary conditions on the 4-leg ladder For the 4-leg ladder, we have also used the Gutzwiller zip- per to obtain the GPFS MPS with periodic boundary condi- 0 N/2 N Subsystem length 𝓁 1 2 3 4 5 6S1(𝓁, N= 4 × 18) GPFS, M = 2400 c ≃ 3.27 GPFS, M = 48...

  89. [97]

    11, we show the 6-leg GPFS MPS and final DMRG state at K/J = 0.6 with different bond dimensions M = 4000, 6000 (cf

    S1 convergence and structure factor data on the 6-leg ladder In Fig. 11, we show the 6-leg GPFS MPS and final DMRG state at K/J = 0.6 with different bond dimensions M = 4000, 6000 (cf. Fig. 5 of the main text). While the final DMRG entanglement entropy is still not fully converg...

  90. [98]

    enhanced

    state seem unlikely (cf. Fig. S11 of Ref. [29]; note that the parameters in that figure correspond to K/J = ∞ in our model—see Sec. VIII of the Supplemental Material of Ref. [29] for a translation of conventions). That is, we know that we can capture entanglement entropy values...

Pith tools

Reviewed August 27, 2026 · model on record in the stance chip above.