The generic Mott transition in the sine-Gordon model through an embedded worm algorithm

arXiv:2510.20901 · cond-mat.str-el · Submitted 2025-10-23 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Quantum Radio. Generated commentary on the latest quantum physics and condensed matter papers.

Kai: I'm Kai.

Mira: And I'm Mira, here with Lev.

Kai: We're looking at the paper: "The generic Mott transition in the sine-Gordon model through an embedded worm algorithm".

Mira: Here's the gist of the paper. The generic Mott transition in one-dimensional quantum systems can be described by the sine-Gordon model with a tilt via bosonization Because the configuration space of the sine-Gordon model separates into distinct topological sectors, standard local Monte Carlo schemes are limited to very small system sizes To overcome this limitation, we introduce the smooth worm (SmoWo) Monte Carlo algorithm which enlarges the configuration space to allow smooth transitions between topological sectors The method combines worm updates with event-chain Monte Carlo moves Thanks to the substantial acceleration achieved by the SmoWo algorithm, we are able to simulate large system sizes, providing a precise picture of the different phases and critical behaviour of the sine-Gordon model.

Kai: Model Formulation and Bosonization The study focuses on the generic Mott transition in 1D quantum systems using a tight-binding model for spinless fermions, which is mapped onto the sine-Gordon model via bosonization At low energies, this involves working with two bosonic fields and identifying particle excitations as kinks or half-kinks in the field representation.

Mira: The resulting bosonized Hamiltonian is given by a sine-Gordon action The path integral formulation of the partition function is expressed in terms of the Euclidean action S

ϕ: = ˆx,τ 1/2πK u(∂xϕ) 2 + 1/u (∂τϕ) 2 - g cos(4ϕ) + µ π/∇ϕ, (8) and the boundary conditions are crucial for recovering correct physics.

Kai: Algorithm Development The standard Metropolis–Hastings algorithm is inefficient because its acceptance rate for topological updates is exponentially small To address this, the Smooth worm algorithm (SmoWo) is introduced, which combines worm updates with local moves of the event-chain Monte Carlo (ECMC) scheme The SmoWo algorithm operates in an enlarged configuration space where topological sectors are smoothly connected.

Title and authors: Kai: Let's talk about who wrote this, the authors of "The generic Mott transition in the sine-Gordon model through an embedded worm algorithm". They are Oscar Bouverot-Dupuis, Laura Foini, and Alberto Rosso. Mira, what's your take on their focus?

Mira: They are focusing on studying the 1D generic Mott transition by mapping it to the sine-Gordon model using bosonization <ref:2510.20901#pg1>. This approach allows them to directly compare their numerical study with analytical results for that model.

Lev: It's interesting that they chose this approach rather than simulating a microscopic quantum Hamiltonian directly, which is what you often see in other quantum Monte Carlo studies.

Kai: They specifically chose the bosonized picture because it serves as a starting point for Monte Carlo simulations of more complicated bosonized systems later on.

Mira: And they work in the grand-canonical ensemble, which complements canonical-ensemble algorithms that we often use.

Lev: That means their findings are relevant to how we approach simulations in these kinds of interacting fermion systems.

Kai: So, they're not just looking at a static property; they're exploring the dynamics across different ensembles to see what happens.

The paper's summary: Kai: So, let's talk about what the paper actually summarizes regarding their findings on "The generic Mott transition in the sine-Gordon model through an embedded worm algorithm". They’re showing how they tackle the problem of sampling those separated topological sectors.

Mira: The summary explains that by using the SmoWo algorithm, which smooths out the configuration space, they can overcome that limitation of standard local Monte Carlo schemes to sample these sectors effectively.

Lev: So they are using this new algorithm to get access to parts of the phase diagram where traditional methods just couldn't go because it would take an exponential amount of time.

Kai: They're showing that the acceleration achieved by SmoWo allows them to simulate large system sizes, which gives them a much more precise picture of the different phases and critical behavior.

Mira: Furthermore, they are providing insights into the phase diagram by looking at scalar observables like doping delta rho, which vanishes in the Mott insulator and concentrates around kappa times mu over R in the Luttinger liquid.

Lev: So they're connecting microscopic excitations to macroscopic properties in a way that’s useful for understanding these transitions.

Kai: And they are also linking the phase-phase two-point function, C theta(p), which is related to the worm configuration, to its time behavior at criticality, suggesting an anomalous dimension of two point nine nine plus or minus zero point zero two <ref:2510.20901#pg2>.

The paper's improvements: Kai: Now let's discuss the specific improvements they propose in "The generic Mott transition in the sine-Gordon model through an embedded worm algorithm". They are highlighting how their SmoWo method is better than the standard Metropolis–Hastings algorithm.

Mira: The main improvement is introducing the smooth worm algorithm itself, which enlarges configuration space to allow smooth transitions between topological sectors, overcoming that limitation of standard local Monte Carlo schemes.

Lev: And they combine that with event-chain Monte Carlo moves to perform non reversible moves, which is what gives it the enhanced sampling efficiency.

Kai: The result of this combination is a substantial acceleration, allowing them to simulate large system sizes and get a precise picture of the sine-Gordon model's phases and critical behavior.

Mira: They also show that this algorithm has an algorithmic dynamical critical exponent z alg point approximately four point eight for the SmoWo algorithm, which is lower than the five point eight you see for the standard worm update algorithm.

Lev: That difference in z alg points is significant because it shows how much better the new method handles those dynamics compared to what you'd expect from a naive approach.

Conclusion: Kai: So, to wrap up on "The generic Mott transition in the sine-Gordon model through an embedded worm algorithm", they proved this SmoWo algorithm is powerful enough to clearly identify the two phases of the model.

Mira: They successfully identify the Luttinger liquid with quasi long range order and the Mott insulator with its characteristic Hubbard bands. They also found critical exponents z equals two point zero zero plus or minus zero point zero one, nu equals zero point five one plus or minus zero point zero one, and eta equals two point nine nine plus or minus zero point zero two.

Lev: Those critical exponents are comparable to analytical results, which is a strong piece of evidence for the physical picture they're describing.

Kai: They also noted that the critical point isn't a Luttinger liquid which has z equals one; instead, the crossover time defines a missing correlation length ξ equal to the particle time tau divided by z which diverges as mu approaches mu minus c.

Mira: And at that transition energy of two mu plus or minus c it increases to a finite value of two mu plus or minus c, and the roles of xi h tau and xi p tau would then be reversed, with xi h tau diverging at the transition.

Lev: It's interesting how they suggest that the Mott insulator can also be exited through negative mu plus or minus c instead of just positive mu plus or minus c.

Kai: They conclude that the sine-Gordon model has two fixed points, but their enlarged model has three, and SmoWo proved powerful enough to clearly identify those two phases.

Mira: So it's a lot of work showing how this algorithm works across different regimes in the paper "The generic Mott transition in the sine-Gordon model through an embedded worm algorithm".

Lev: For me, it just shows that for complex bosonized systems, having a clever sampling method can make the difference between getting stuck or actually seeing the physics.

Universit´e Paris Saclay · IPhT · CNRS

cond-mat.str-el

Submitted: 2025-10-23

Updated: 2026-11-05

Comments: 35 pages, 20 figures

DOI: 10.1103/4k6t-lnzp

Code: https://github.com/OscarBouverot-Dupuis/algo-phys

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 76/100

The gist: The generic Mott transition in one-dimensional quantum systems can be described by the sine-Gordon model with a tilt via bosonization Because the configuration space of the sine-Gordon model…

Key concepts

Sine-Gordon Model
This is a mathematical model used to describe generic Mott transitions in one-dimensional quantum systems. It arises from mapping the tight-binding model of spinless fermions onto two bosonic fields, where particle excitations are represented as kinks or half-kinks in the field representation.
Smooth Worm (SmoWo) Algorithm
This is an advanced Monte Carlo simulation technique designed to sample the sine-Gordon model efficiently. It combines traditional worm updates with Event-Chain Monte Carlo moves to smoothly connect different topological sectors of the configuration space, allowing simulations on large system sizes.
Luttinger Liquid (LL) and Mott Insulator (MI)
These are the two distinct phases identified in the model's phase diagram. The LL exhibits quasi-long-range order, while the MI is characterized by characteristic Hubbard bands. Observables like doping ($ ho$) help distinguish between these phases.
Critical Exponents
These are specific numbers (like z=2.00 and eta=2.99) that describe how physical quantities behave near a critical point in the system. The SmoWo algorithm successfully retrieved these values, confirming the model's critical behavior against analytical predictions.

Terminology

Summary

The generic Mott transition in one-dimensional quantum systems can be described by the sine-Gordon model with a tilt via bosonization Because the configuration space of the sine-Gordon model separates into distinct topological sectors, standard local Monte Carlo schemes are limited to very small system sizes To overcome this limitation, we introduce the smooth worm (SmoWo) Monte Carlo algorithm which enlarges the configuration space to allow smooth transitions between topological sectors The method combines worm updates with event-chain Monte Carlo moves Thanks to the substantial acceleration achieved by the SmoWo algorithm, we are able to simulate large system sizes, providing a precise picture of the different phases and critical behaviour of the sine-Gordon model. p. 0

Where the paper says it

The generic Mott transition in the sine-Gordon model through an embedded worm algorithm Oscar Bouverot-Dupuis,1, 2 Laura Foini,2 and Alberto Rosso1 1Universit´e Paris Saclay, CNRS, LPTMS, 91405, Orsay, France 2 IPhT, CNRS, CEA, Universit´e Paris Saclay, 91191 Gif-sur-Yvette, France (Dated: October 27, 2025) The generic Mott transition in one-dimensional quantum systems can be described by the sineGordon model with a tilt via bosonization. Because the configuration space of the sine-Gordon model separates into distinct topological sectors, standard local Monte Carlo schemes are limited to very small system sizes.

Model Formulation and Bosonization

The study focuses on the generic Mott transition in 1D quantum systems using a tight-binding model for spinless fermions, which is mapped onto the sine-Gordon model via bosonization At low energies, this involves working with two bosonic fields and identifying particle excitations as kinks or half-kinks in the field representation. The resulting bosonized Hamiltonian is given by a sine-Gordon action The path integral formulation of the partition function is expressed in terms of the Euclidean action S[ϕ] = ˆx,τ 1/2πK u(∂xϕ) 2 + 1/u (∂τϕ) 2 - g cos(4ϕ) + µ π/∇ϕ, (8) and the boundary conditions are crucial for recovering correct physics. p. 2

Where the paper says it

In the path-integral formalism, integrating out the field θ shows that the grand-canonical equilibrium partition function Z = Tr e −βHˆ at inverse temperature β = 1/T can be expressed as Z = X +∞ Nx,Nτ =−∞ ˆ Dϕ ϕ(0,τ)=ϕ(L,τ)+πNx ϕ(x,0)=ϕ(x,β)+πNτ e −S[ϕ] , (7) where the Euclidean action is the sine-Gordon action S[ϕ] = ˆ x,τ 1 2πK u(∂xϕ) 2 + 1 u (∂τϕ) 2 − g cos(4ϕ) + µ π ∂xϕ, (8) with the notation ´ x,τ = ´ L 0 dx ´ β 0 dτ . The boundary conditions in the path integral (7) are crucial to recover the correct physics.

Algorithm Development

The standard Metropolis–Hastings algorithm is inefficient because its acceptance rate for topological updates is exponentially small To address this, the Smooth worm algorithm (SmoWo) is introduced, which combines worm updates with local moves of the event-chain Monte Carlo (ECMC) scheme The SmoWo algorithm operates in an enlarged configuration space where topological sectors are smoothly connected. The ECMC algorithm is a class of continuous-time and rejection-free algorithms that perform non-reversible moves, which are used to smooth the worm throughout its construction. p. 5

Where the paper says it

To overcome this issue, we propose a Worm algorithm (Wo) and its enhanced version, the Smooth Worm algorithm (SmoWo), which perform two types of updates: • local ϕ updates at constant Nx and Nτ using the Event-Chain Monte Carlo (ECMC) algorithm, • Nx and Nτ updates using the worm algorithm, enhanced with ECMC moves for the SmoWo algorithm. In a nutshell, ECMC algorithms [27, 28] are a class of continuous-time and rejection-free algorithms that are non-reversible as they do not satisfy detailed balance but only the weaker global balance.

Performance and Critical Exponents

The performance of the SmoWo algorithm is assessed by calculating the integrated autocorrelation time, which scales as O(L 4.8) for a system of size ∼ L × L squared This scaling is an improvement over the exponential scaling of the naive Metropolis algorithm. p. 7

Where the paper says it

Although this value of zalg. is quite high, the polynomial scaling of this algorithm is a dramatic improvement over the exponential scaling of the naive Metropolis algorithm described in Sec.

The results show that the algorithmic dynamical critical exponent is zalg. p. 7
Where the paper says it

It appears that for both observables τint ∼ L zalg. 8 the algorithmic dynamical exponent (not to be confused with the dynamical critical exponent z).

≃ 4.8 for the SmoWo algorithm, which is a decrease from zalg. p. 8
Where the paper says it

The net result is a decrease of the algorithmic dynamical critical exponent from zalg. 8 for the Wo algorithm to zalg.

≃ 5.8 for the Wo algorithm

Physical Insights and Observables

The phase diagram shows two distinct phases: the Luttinger liquid (LL) with quasi-long range order and the Mott insulator (MI) with its characteristic Hubbard bands Scalar observables include the doping δρ ≡ Nx/L, which vanishes in the MI and concentrates around KRµR in the LL. The phase-phase two-point function Cθ(p) is linked to the worm configuration, and its time behavior at criticality suggests an anomalous dimension η = 2. p. 0

Where the paper says it

Scalar observables 10 C. Phase-phase two-point function 11 D.

99 ± 0.02.

Conclusion

The SmoWo algorithm proves powerful enough to clearly identify the two phases of the model and retrieve critical exponents z = 2.00 ± 0.01, ν = 0.51 ± 0.01 and η = 2.99 ± 0.02, all of which compare favourably with analytical results The algorithm is relevant for studying any bosonized system in the grand-canonical ensemble. p. 1

Where the paper says it

This enables direct comparison with analytical results on the same model and serves as a starting point for Monte Carlo simulations of more complicated bosonized systems. We work in the grand-canonical ensemble, thereby complementing the canonical-ensemble algorithm of Ref.

The SmoWo algorithm has a decrease of the algorithmic dynamical critical exponent from zalg. p. 8
Where the paper says it

The net result is a decrease of the algorithmic dynamical critical exponent from zalg. 8 for the Wo algorithm to zalg.

≃ 5.8 for the Wo algorithm to zalg. p. 7
Where the paper says it

Although this value of zalg. is quite high, the polynomial scaling of this algorithm is a dramatic improvement over the exponential scaling of the naive Metropolis algorithm described in Sec.

≃ 4.8 for the SmoWo algorithm. p. 0
Where the paper says it

Because the configuration space of the sine-Gordon model separates into distinct topological sectors, standard local Monte Carlo schemes are limited to very small system sizes. To overcome this limitation, we introduce the smooth worm (SmoWo) Monte Carlo algorithm which enlarges the configuration space to allow smooth transitions between topological sectors.

The results show that the sine-Gordon model has two fixed points, but the enlarged model has three. p. 0
Where the paper says it

Analytically, apart from strongcoupling expansions of the Bose Hubbard model [9], most results rely on the description of the transition through the sine–Gordon model via bosonization [10–14]. It was thus established that 1D Mott transitions fall into two distinct universality classes: varying the interaction strength at fixed commensurate densities leads to a Berezinskii–Kosterlitz– arXiv:2510.

The smoothed worm algorithm proved powerful enough to clearly identify the two phases: the Luttinger liquid with quasi-long range order, and the Mott insulator with its characteristic Hubbard bands. p. 9
Where the paper says it

Phase diagram of the Mott transition. As g is increased, the lobe of the Mott insulator (MI) eats on the Luttinger liquid (LL).

The critical point is not a Luttinger liquid which has z = 1. p. 9
Where the paper says it

This shows that the topological variables Nx, Nτ decouple from the periodic field φ. The action for φ is that of a Luttinger liquid (LL).

p. 9
Where the paper says it

This shows that the topological variables Nx, Nτ decouple from the periodic field φ. The action for φ is that of a Luttinger liquid (LL).

The crossover time ξpτ — the superscript p refers to Cθ(0, τ) describing the propagation of a particle — defines the previously missing correlation length ξ = (ξpτ) 1/z which diverges as µ → µ−c. The energy of a hole being E1h = µ + µc, it increases to the finite value 2µc at the transition. The roles of ξhτ and ξpτ would then be reversed, with ξhτ diverging at the transition. p. 0
Where the paper says it

The generic Mott transition in the sine-Gordon model through an embedded worm algorithm Oscar Bouverot-Dupuis,1, 2 Laura Foini,2 and Alberto Rosso1 1Universit´e Paris Saclay, CNRS, LPTMS, 91405, Orsay, France 2 IPhT, CNRS, CEA, Universit´e Paris Saclay, 91191 Gif-sur-Yvette, France (Dated: October 27, 2025) The generic Mott transition in one-dimensional quantum systems can be described by the sineGordon model with a tilt via bosonization. Because the configuration space of the sine-Gordon model separates into distinct topological sectors, standard local Monte Carlo schemes are limited to very small system sizes.

p. 0
Where the paper says it

The generic Mott transition in the sine-Gordon model through an embedded worm algorithm Oscar Bouverot-Dupuis,1, 2 Laura Foini,2 and Alberto Rosso1 1Universit´e Paris Saclay, CNRS, LPTMS, 91405, Orsay, France 2 IPhT, CNRS, CEA, Universit´e Paris Saclay, 91191 Gif-sur-Yvette, France (Dated: October 27, 2025) The generic Mott transition in one-dimensional quantum systems can be described by the sineGordon model with a tilt via bosonization. Because the configuration space of the sine-Gordon model separates into distinct topological sectors, standard local Monte Carlo schemes are limited to very small system sizes.

Appendix Details

The appendix details the bosonization approach used in this work, including bosonization conventions and the path integral formulation of the partition function It also provides a proof of the validity of the algorithm and implementation details. The appendix also contains pseudocode implementations for the Wo and SmoWo algorithms. p. 8

Where the paper says it

A proof of the validity of the algorithm and a pseudocode implementation are given respectively in Appendix B and C 2. Performance wise, the integrated autocorrelation time of the SmoWo algorithm is given in Fig.

The appendix computes various event times associated to the quadratic interactions, cosine interaction, refreshment term, and worm events. p. 6
Where the paper says it

For the algorithm described above, there are four types of event times to compute when moving fi along e: the quadratic ones t i,j q , the cosine one t i c , the refreshment one tr and the worm one tw. These event times can be explicitly computed using inversion sampling (see Appendix C 1).

p. 6
Where the paper says it

For the algorithm described above, there are four types of event times to compute when moving fi along e: the quadratic ones t i,j q , the cosine one t i c , the refreshment one tr and the worm one tw. These event times can be explicitly computed using inversion sampling (see Appendix C 1).

p. 6
Where the paper says it

For the algorithm described above, there are four types of event times to compute when moving fi along e: the quadratic ones t i,j q , the cosine one t i c , the refreshment one tr and the worm one tw. These event times can be explicitly computed using inversion sampling (see Appendix C 1).

It also provides a detailed proof of the global balance condition for the SmoWo algorithm. p. 5
Where the paper says it

To overcome this issue, we propose a Worm algorithm (Wo) and its enhanced version, the Smooth Worm algorithm (SmoWo), which perform two types of updates: • local ϕ updates at constant Nx and Nτ using the Event-Chain Monte Carlo (ECMC) algorithm, • Nx and Nτ updates using the worm algorithm, enhanced with ECMC moves for the SmoWo algorithm. In a nutshell, ECMC algorithms [27, 28] are a class of continuous-time and rejection-free algorithms that are non-reversible as they do not satisfy detailed balance but only the weaker global balance.

Numerical Results Summary

The numerical results show that for µ 0 and δρ = κ = ρs = 0, meaning the field ϕ gets trapped in one minimum and kinks are suppressed: this the MI For µ > µc, the field ϕ is tilted since δρ > 0 and fluctuates a lot since κ, ρs > 0 and C2kF is small (and decreases with system size): this is the LL. The transition at the tip of the lobe belongs to the BKT (Berezinskii–Kosterlitz–Thouless) universality class. p. 9

Where the paper says it

In this work, we focus on the generic transition which is crossed by varying µ as shown by the red arrow. The transition at the tip of the lobe is different as it belongs to the BKT (Berezinskii–Kosterlitz–Thouless) universality class [2, 3, 49].

The finite-size scaling collapse for κ and ρs shows that the two scaling functions for κ and ρs are the same up to a linear rescaling of the x and y axes. p. 5
Where the paper says it

12 × 10−2 ) β (and similarly for β ↔ L). This is a ridiculously small acceptance rate which, along with its exponential scaling, should discourage anyone from trying to attempt large system-size simulations with the Metropolis–Hastings algorithm.

The data suggests that there is no diverging correlation length when coming from the MI and that, right at the critical point, the system is still a MI. p. 7
Where the paper says it

(26) We extract from it the integrated autocorrelation time τ O int = 1 2 P+∞ t=−∞ CO(t) which is the time needed to generate a new independent sample once the Markov process has thermalized (or mixed [45]). To assess the critical slowing-down of the algorithm, we perform this analysis at the critical point of the generic Mott transition by scaling β ∼ L 2 since the dynamical critical exponent is z = 2 (see Sec.

The analysis reveals that the sine-Gordon model has two fixed points, but the enlarged model has three. p. 0
Where the paper says it

Analytically, apart from strongcoupling expansions of the Bose Hubbard model [9], most results rely on the description of the transition through the sine–Gordon model via bosonization [10–14]. It was thus established that 1D Mott transitions fall into two distinct universality classes: varying the interaction strength at fixed commensurate densities leads to a Berezinskii–Kosterlitz– arXiv:2510.

The SmoWo algorithm proved powerful enough to clearly identify the two phases: the Luttinger liquid with quasi-long range order, and the Mott insulator with its characteristic Hubbard bands. p. 9
Where the paper says it

Phase diagram of the Mott transition. As g is increased, the lobe of the Mott insulator (MI) eats on the Luttinger liquid (LL).

The critical point is not a Luttinger liquid which has z = 1. The crossover time ξpτ — the superscript p refers to Cθ(0, τ) describing the propagation of a particle — defines the previously missing correlation length ξ = (ξpτ) 1/z which diverges as µ → µ−c. The energy of a particle and a hole is E1p1h = E1p + E1h = 2µc, which, as expected, is µ-independent. p. 2
Where the paper says it

An extra hole (or spin down) creates a +π-kink. Middle: With a background density of 1 particle every two sites, a particle (hole) can decay into two half-particles (halfholes) corresponding to two consecutive particles (holes).

The MI can also be exited through −µc instead of +µc. p. 9
Where the paper says it

(31) This is the Mott insulator (MI). We thus expect the existence of two critical points ±µc (see Fig.

The roles of ξhτ and ξpτ would then be reversed, with ξhτ diverging at the transition.

References

[1] N. F. Mott, Metal-Insulator Transition, Rev. p. 0

Where the paper says it

Pseudocode implementations 32 References 34 I. INTRODUCTION Over the past decades, the Mott transition [1– 4] describing metal-insulator transitions driven by strong electronic interactions has attracted a great deal of interest.

Mod. Phys. 40, 677 (1968) [2] M. P. A. Fisher, P. B. Weichman, G. Grinstein, and D. S. Fisher, Boson localization and the superfluidinsulator transition, Phys Rev B 40, 546 (1989) [3] T. Giamarchi, Mott transition in one dimension, Physica B: Condensed Matter 230-232, 975 (1997), proceedings of the International Conference on Strongly Correlated Electron Systems [4] N. p. 0
Where the paper says it

The generic Mott transition in the sine-Gordon model through an embedded worm algorithm Oscar Bouverot-Dupuis,1, 2 Laura Foini,2 and Alberto Rosso1 1Universit´e Paris Saclay, CNRS, LPTMS, 91405, Orsay, France 2 IPhT, CNRS, CEA, Universit´e Paris Saclay, 91191 Gif-sur-Yvette, France (Dated: October 27, 2025) The generic Mott transition in one-dimensional quantum systems can be described by the sineGordon model with a tilt via bosonization. Because the configuration space of the sine-Gordon model separates into distinct topological sectors, standard local Monte Carlo schemes are limited to very small system sizes.

Mott, Metal-insulator transitions (CRC Press, 2004) [5] A. p. 0
Where the paper says it

Pseudocode implementations 32 References 34 I. INTRODUCTION Over the past decades, the Mott transition [1– 4] describing metal-insulator transitions driven by strong electronic interactions has attracted a great deal of interest.

Georges, G. Kotliar, W. Krauth, and M. J. Rozenberg, Dynamical mean-field theory of strongly correlated fermion systems and the limit of infinite dimensions, Rev. Mod. Phys. 68, 13 (1996) [6] M. Imada, A. Fujimori, and Y. Tokura, Metalinsulator transitions, Rev. p. 0
Where the paper says it

Because the configuration space of the sine-Gordon model separates into distinct topological sectors, standard local Monte Carlo schemes are limited to very small system sizes. To overcome this limitation, we introduce the smooth worm (SmoWo) Monte Carlo algorithm which enlarges the configuration space to allow smooth transitions between topological sectors.

Mod. Phys. 70, 1039 (1998) [7] G. G. Batrouni, R. T. Scalettar, and G. T. Zimanyi, Quantum critical phenomena in one-dimensional Bose systems, Phys Rev Lett 65, 1765 (1990) [8] T. p. 0
Where the paper says it

While in dimensions greater than one, a comprehensive theoretical framework remains elusive despite significant numerical efforts [5, 6], the one-dimensional (1D) case has seen major advances. Numerically, the Bose Hubbard model has become a paradigmatic model to observe the Mott transition with quantum Monte Carlo algorithms [7] and the density matrix renormalization group (DMRG) [8].

D. K¨uhner, S. R. White, and H. Monien, Onedimensional Bose-Hubbard model with nearestneighbor interaction, Phys Rev B 61, 12474 (2000) [9] N. p. 0
Where the paper says it

Analytically, apart from strongcoupling expansions of the Bose Hubbard model [9], most results rely on the description of the transition through the sine–Gordon model via bosonization [10–14]. It was thus established that 1D Mott transitions fall into two distinct universality classes: varying the interaction strength at fixed commensurate densities leads to a Berezinskii–Kosterlitz– arXiv:2510.

Elstner and H. Monien, Dynamics and thermodynamics of the Bose-Hubbard model, Phys Rev B 59, 12184 (1999) [10] F. p. 0
Where the paper says it

While in dimensions greater than one, a comprehensive theoretical framework remains elusive despite significant numerical efforts [5, 6], the one-dimensional (1D) case has seen major advances. Numerically, the Bose Hubbard model has become a paradigmatic model to observe the Mott transition with quantum Monte Carlo algorithms [7] and the density matrix renormalization group (DMRG) [8].

D. M. Haldane, Effective Harmonic-Fluid Approach to Low-Energy Properties of OneDimensional Quantum Fluids, Phys Rev Lett 47, 1840 (1981) [11] T. Giamarchi, Quantum physics in one dimension, International series of monographs on physics (Clarendon Press, Oxford, 2004) [12] J. p. 2
Where the paper says it

In the path-integral formalism, integrating out the field θ shows that the grand-canonical equilibrium partition function Z = Tr e −βHˆ at inverse temperature β = 1/T can be expressed as Z = X +∞ Nx,Nτ =−∞ ˆ Dϕ ϕ(0,τ)=ϕ(L,τ)+πNx ϕ(x,0)=ϕ(x,β)+πNτ e −S[ϕ] , (7) where the Euclidean action is the sine-Gordon action S[ϕ] = ˆ x,τ 1 2πK u(∂xϕ) 2 + 1 u (∂τϕ) 2 − g cos(4ϕ) + µ π ∂xϕ, (8) with the notation ´ x,τ = ´ L 0 dx ´ β 0 dτ . The boundary conditions in the path integral (7) are crucial to recover the correct physics.

von Delft and H. Schoeller, Bosonization for beginners — refermionization for experts, Annalen der Physik 510, 225 (1998) [13] M. A Cazalilla, Bosonizing one-dimensional cold atomic gases, Journal of Physics B: Atomic, Molecular and Optical Physics 37, S1–S47 (2004) [14] N. p. 2
Where the paper says it

In the path-integral formalism, integrating out the field θ shows that the grand-canonical equilibrium partition function Z = Tr e −βHˆ at inverse temperature β = 1/T can be expressed as Z = X +∞ Nx,Nτ =−∞ ˆ Dϕ ϕ(0,τ)=ϕ(L,τ)+πNx ϕ(x,0)=ϕ(x,β)+πNτ e −S[ϕ] , (7) where the Euclidean action is the sine-Gordon action S[ϕ] = ˆ x,τ 1 2πK u(∂xϕ) 2 + 1 u (∂τϕ) 2 − g cos(4ϕ) + µ π ∂xϕ, (8) with the notation ´ x,τ = ´ L 0 dx ´ β 0 dτ . The boundary conditions in the path integral (7) are crucial to recover the correct physics.

Dupuis, Field Theory of Condensed Matter and Ultracold Gases, Vol. p. 0
Where the paper says it

The generic Mott transition in the sine-Gordon model through an embedded worm algorithm Oscar Bouverot-Dupuis,1, 2 Laura Foini,2 and Alberto Rosso1 1Universit´e Paris Saclay, CNRS, LPTMS, 91405, Orsay, France 2 IPhT, CNRS, CEA, Universit´e Paris Saclay, 91191 Gif-sur-Yvette, France (Dated: October 27, 2025) The generic Mott transition in one-dimensional quantum systems can be described by the sineGordon model with a tilt via bosonization. Because the configuration space of the sine-Gordon model separates into distinct topological sectors, standard local Monte Carlo schemes are limited to very small system sizes.

2 (unpublished) (WORLD SCIENTIFIC (EUROPE), 2026) see https://www. p. 2
Where the paper says it

by a downward kink of amplitude π (see Fig. 1, top) since ˆ L 0 dx(ˆρ(x) − ρ0) = ϕˆ(0) − ϕˆ(L) π .

lptmc.jussieu.fr/users/dupuis [15] V. p. 0
Where the paper says it

The generic Mott transition in the sine-Gordon model through an embedded worm algorithm Oscar Bouverot-Dupuis,1, 2 Laura Foini,2 and Alberto Rosso1 1Universit´e Paris Saclay, CNRS, LPTMS, 91405, Orsay, France 2 IPhT, CNRS, CEA, Universit´e Paris Saclay, 91191 Gif-sur-Yvette, France (Dated: October 27, 2025) The generic Mott transition in one-dimensional quantum systems can be described by the sineGordon model with a tilt via bosonization. Because the configuration space of the sine-Gordon model separates into distinct topological sectors, standard local Monte Carlo schemes are limited to very small system sizes.

L. Pokrovsky and A. L. Talapov, Ground State, Spectrum, and Phase Diagram of Two-Dimensional Incommensurate Crystals, Phys Rev Lett 42, 65 (1979) [16] H. J. Schulz, Critical behavior of commensurateincommensurate phase transitions in two dimensions, Phys Rev B 22, 5274 (1980) [17] F. p. 5
Where the paper says it

The worm algorithm [25, 26] can be applied to any model whose configurations are made of loops, which for the sine-Gordon model we have identified in the current-fluctuation representation (16). By allowing intermediate configurations to have one open path (the “worm”), the worm algorithm drastically reduces the critical slowing down observed near phase transitions.

D. M. Haldane, P. Bak, and T. Bohr, Phase diagrams of surface structures from Bethe-ansatz solutions of the quantum sine-Gordon model, Phys Rev B 28, 2743 (1983) [18] V. p. 1
Where the paper says it

The latter is also well-known for describing commensurate-incommensurate transitions in uniaxial surface structures [15–17] and related systems [18, 19]. Even within the sine-Gordon picture, however, its analysis proved non-trivial as a complete picture of the transition only came through the combination of many integrability [17, 20, 21] and fieldtheoretical techniques [3, 16, 22, 23].

L. Pokrovsky and A. Virosztek, Long-range interactions in commensurate-incommensurate phase transition, Journal of Physics C: Solid State Physics 16, 4513 (1983) [19] L. p. 1
Where the paper says it

Thouless (BKT) transition, while doping the system gives rise to the generic (or Mott-δ) transition. The latter is also well-known for describing commensurate-incommensurate transitions in uniaxial surface structures [15–17] and related systems [18, 19].

Radzihovsky, Thermal depinning and transversefield tilting transitions in a planar vortex array pinned by a columnar defect, Phys Rev B 73, 104504 (2006) [20] H. Yamamoto, Commensurate-Incommensurate Transition in a Two Dimensional Classical SineGordon System, Progress of Theoretical Physics 69, 1291 (1983) [35] M. p. 1
Where the paper says it

Thouless (BKT) transition, while doping the system gives rise to the generic (or Mott-δ) transition. The latter is also well-known for describing commensurate-incommensurate transitions in uniaxial surface structures [15–17] and related systems [18, 19].

Wallin, E. S. Sørensen, S. M. Girvin, and A. P. Young, Superconductor-insulator transition in twodimensional dirty boson systems, Phys Rev B 49, 12115 (1994) [36] N Metropolis, A W Rosenbluth, M N Rosenbluth, A H Teller, and E Teller, Equation of State Calculations by Fast Computing Machines The Journal of Chemical Physics 21, 1087 (1953) [37] R. H. Swendsen and J.-S. Wang, Nonuniversal critical dynamics in Monte Carlo simulations Phys Rev Lett 58, 86 (1987) [38] U Wolff, Collective Monte Carlo Updating for Spin Systems Phys Rev Lett 62, 361 (1989) [39] M Hasenbusch, M Marcu, and K Pinn, The sine Gordon model: perturbation theory and cluster Monte Carlo Physica A: Statistical Mechanics and its Applications 211, 255 (1994) [40] P Werner and M Troyer, Efficient Simulation of Resistively Shunted Josephson Junctions Phys Rev Lett 95, 060201 (2005) [41] A Monemvassitis, A Guillin, and M Michel, PDMP characterisation of event-chain Monte Carlo algorithms for particle systems Journal of Statistical Physics 190, 66 (2023) [42] Y. p. 5
Where the paper says it

Indeed, cluster algorithms used for bosonized systems [24, 39, 40] cannot change the boundary conditions, and, furthermore, crucially rely on the particle-hole symmetry ϕ → −ϕ which is broken by the chemical potential µ. To overcome this issue, we propose a Worm algorithm (Wo) and its enhanced version, the Smooth Worm algorithm (SmoWo), which perform two types of updates: • local ϕ updates at constant Nx and Nτ using the Event-Chain Monte Carlo (ECMC) algorithm, • Nx and Nτ updates using the worm algorithm, enhanced with ECMC moves for the SmoWo algorithm.

Nishikawa, M Michel, W Krauth, and K. Hukushima Event-chain algorithm for the Heisenberg model: Evidence for z ≃ 1 dynamic scaling Phys Rev E 92, 063306 (2015) [43] M Michel, J Mayer, and W Krauth, Event-chain Monte Carlo for classical continuous spin models EPL (Europhysics Letters) 112, 20003 (2015) [44] M Klement, S Lee, J A Anderson, and M Engel Newtonian Event-Chain Monte Carlo and Collision Prediction with Polyhedral Particles Journal of Chemical Theory and Computation 17, 4686 (2021) [45] D A Levin and Y Peres, Markov chains and mixing times Vol. p. 1
Where the paper says it

The algorithm combines updates from the worm algorithm [25, 26] with local moves of the event-chain Monte Carlo (ECMC) scheme [27, 28]. The worm algorithm is designed to sample loop models and has already proven effective to study Mott transitions in coarse-grained (2 + 1)D models [29– 32].

107 (American Mathematical Soc. p. 7
Where the paper says it

The difference between both scalings is traced back to the fact that the worm algorithm can reach many configurations and therefore select a typical one, while the rigid Metropolis–Hastings updates only propose very atypical configurations. 20 40 60 80 100 L 104 105 106 107 τint (sweeps) L 4.

, 2017) [46] O F Syljurasen and A W Sandvik, Quantum Monte Carlo with directed loops Phys Rev E 66, 046701 (2002) [47] M A Novotny, Monte Carlo Algorithms with Absorbing Markov Chains: Fast Local Algorithms for Slow Dynamics Phys Rev Lett 74, 1 (1995) [48] A Bortz, M Kalos, and J Lebowitz, A new algorithm for Monte Carlo simulation of Ising spin systems Journal of Computational Physics 17, 10 (1975) [49] R Daviet and N Dupuis, Nonperturbative Functional Renormalization-Group Approach to the Sine-Gordon Model and the LukyanovZamolodchikov Conjecture Phys Rev Lett 122, 155301 (2019] [50] This is analogous to what is done for the Ising model. p. 0
Where the paper says it

Numerically, the Bose Hubbard model has become a paradigmatic model to observe the Mott transition with quantum Monte Carlo algorithms [7] and the density matrix renormalization group (DMRG) [8]. Analytically, apart from strongcoupling expansions of the Bose Hubbard model [9], most results rely on the description of the transition through the sine–Gordon model via bosonization [10–14].

Instead of computing the magnetization m, one usually considers its absolute value m [60] [51] A M Ferrenberg and R H Swendsen, New Monte Carlo technique for studying phase transitions Phys Rev Lett 61, 2635 (1988) [52] E Orignac, One-dimensional Bose-Hubbard model with long-range hopping Phys Rev B 112, 045424 (2025] [53] E Malatsetxebarria, Z Cai, U Schollw¨ock, and M A Cazalilla Dissipative effects on the superfluidto-insulator transition in mixed-dimensional optical lattices Phys Rev A 88, 063630 (2013] [54] N Paris, N Dupuis, and C Mora Three-channel charge Kondo model at high transparency (2025), arXiv:2509. p. 0
Where the paper says it

The generic Mott transition in the sine-Gordon model through an embedded worm algorithm Oscar Bouverot-Dupuis,1, 2 Laura Foini,2 and Alberto Rosso1 1Universit´e Paris Saclay, CNRS, LPTMS, 91405, Orsay, France 2 IPhT, CNRS, CEA, Universit´e Paris Saclay, 91191 Gif-sur-Yvette, France (Dated: October 27, 2025) The generic Mott transition in one-dimensional quantum systems can be described by the sineGordon model with a tilt via bosonization. Because the configuration space of the sine-Gordon model separates into distinct topological sectors, standard local Monte Carlo schemes are limited to very small system sizes.

03612 [54] [55] https://github.com/OscarBouverot-Dupuis/algo-phys/tree/master/

Appendix A Details

In this appendix, we explain in greater detail the bosonization approach used in this work, including bosonization conventions and the path integral formulation of the phase-phase two-point function It also provides a proof of the algorithm's validity and implementation details.

Appendix B Details

The appendix details how to derive the infinitesimal generator of an ECMC algorithm from a simple example It also provides pseudocode implementations of the Wo and SmoWo algorithms. The appendix computes various event times associated to the quadratic interactions, cosine interaction, refreshment term, and worm events.

Appendix C Details

This section provides detailed pseudocode implementations of the Wo algorithm (Alg. 2) and SmoWo algorithm (Alg. p. 0

Where the paper says it

Because the configuration space of the sine-Gordon model separates into distinct topological sectors, standard local Monte Carlo schemes are limited to very small system sizes. To overcome this limitation, we introduce the smooth worm (SmoWo) Monte Carlo algorithm which enlarges the configuration space to allow smooth transitions between topological sectors.

3) presented in Section III It also provides pseudocode implementations of the computation of observables. p. 1
Where the paper says it

Section III introduces the SmoWo algorithm and evaluates its performance. Section IV defines the observables of interest and outlines their computation within the SmoWo algorithm.

The appendix computes various event times associated to the quadratic interactions, cosine interaction, refreshment term, and worm events.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. SmoWo algorithm implementation for generic Mott transitions: The improved system can simulate large system sizes by using an enlarged configuration space that allows smooth transitions between topological sectors, overcoming the limitation of standard local Monte Carlo schemes.

  2. Enhanced sampling efficiency via ECMC integration: The AI can utilize Event-Chain Monte Carlo moves to perform persistent and non-reversible updates, which are significantly faster than classical reversible methods like Metropolis-Hastings for sampling sine-Gordon models.

  3. Identification of phase behavior through observables: The system can accurately identify the two phases (Luttinger liquid and Mott insulator) by monitoring scalar observables such as the doping delta rho, the amplitude of the 2kF density modulation C2kF, and superfluid stiffness rho s.

  4. Critical exponent determination: The improved system can retrieve critical exponents z = 2.00 +/- 0.01, nu = 0.51 +/- 0.01, and the anomalous dimension for the particle-particle two-point function eta = 2.99 +/- 0.02 by applying finite-size scaling collapse analysis to numerical data.

  5. Renormalization group flow mapping: The AI can map the schematic renormalization group flow of the enlarged model, showing trajectories that move from the Mott insulator fixed point to a continuum of Luttinger liquid fixed points, revealing a lack of an intermediate critical fixed point.

  6. Calculation of phase-phase correlations: The system can calculate the phase-phase two-point function C theta(0, tau) by utilizing the worm algorithm to sample configurations with a worm ranging from particle time pt to hole time ph, providing insights into the algebraic decay in different phases.

  7. Identification of Hubbard bands: The AI can distinguish between the Mott insulator and Luttinger liquid states by analyzing the two-point function C theta(0, tau), where the presence or absence of this correlation distinguishes between excitations in the upper and lower Hubbard bands.

Abstract

The generic Mott transition in one-dimensional quantum systems can be described by the sine-Gordon model with a tilt via bosonization. Because the configuration space of the sine-Gordon model separates into distinct topological sectors, standard local Monte Carlo schemes are limited to very small system sizes. To overcome this limitation, we introduce the smooth worm (SmoWo) Monte Carlo algorithm which enlarges the configuration space to allow smooth transitions between topological sectors. The method combines worm updates with event-chain Monte Carlo moves. We explicitly prove its validity and quantify its performance. Thanks to the substantial acceleration achieved by the SmoWo algorithm, we are able to simulate large system sizes, providing a precise picture of the different phases and critical behaviour of the sine-Gordon model.

Sources

Related papers