betaFeedback

WorksPhysics and Astronomy

Dressed higher-form symmetries in mixed state quantum matter

Abstract

We describe how to construct emergent strong higher-form symmetries in mixed quantum states that act unitarily on Hilbert space. Our construction uses (quasi-)local recovery channels from quantum error correction to some nearby simple (stabilizer) model. We prove that when the recovery is done in an appropriate (deterministic) manner, the resulting operators continue to be unitary. When applied to operators acting on topologically non-trivial sub-manifolds (loops, surfaces), the resulting unitaries inherit the logical algebra of the fixed point model. More generally, the unitarity of the emergent symmetries allows one to define emergent excitations whose statistics follow those of the fixed point limit.

Introduction

Much of our understanding of phases of matter is organized around the notion of symmetry. This idea has seen renewed enthusiasm in recent years with the realization that even many of the phases that were believed to go beyond the usual notion of spontaneous symmetry breaking (notably, topological order) can in fact be understood in this framework by appropriately broadening the notion of symmetries. Particularly important in this context are so-called higher-form symmetries, which act not on the whole system, but on lower-dimensional sub-manifolds [1, 2]. The characteristic (e.g. anyon) excitations of topologically ordered systems arise as the symmetry defects of such higher-form symmetries and their exchange and braiding statistics encode anomalies associated to these symmetries [3].

A key difference between usual (0-form) symmetry breaking and its higher form generalizations is the enhanced robustness of the latter. While usual Landau symmetry breaking phases are only well-defined if the system obeys an exact symmetry condition, topological order tends to be robust against arbitrary local perturbations [4]. This can be understood through the notion of emergent symmetries. This notion is made precise in the context of ground state phases of matter through quasi-adiabatic continuation, which constructs a quasi-local unitary operator relating ground states of the perturbed model to those of the original [5, 6]. The same unitary can be used to “smear out” the higher-form symmetry operators such that these dressed symmetries will leave the new ground state space invariant. The resulting “dressed” symmetries have similar (albeit slightly fattened) supports and have the same algebraic relations as the originals, thus inheriting e.g. the braiding statistics of the excitations they create.

The above picture applies to zero temperature pure ground states. In contrast, mixed states of quantum many-body systems arise naturally once we go from zero to finite temperatures and in more general open quantum systems as a result of (natural or engineered) dissipation. Understanding phases of matter, and particularly topological order, in such mixed quantum many-body states is of much current interest [7, 8, 9, 10]. Symmetries again play a crucial role, with the added twist that there are now two distinct kinds: weak and strong [11]. Anomalies of strong higher-form symmetries play a particularly important role [3, 12, 13]. An obvious question in this context concerns the role of emergent symmetries. Suppose a given state has known symmetry generators, and you now perturb it (e.g. by raising its temperature or by applying some local noise to it). Does the resulting state—which is in general mixed—inherit the symmetries of the original?

There has been important progress on this question [12, 14, 15, 16] (see Sec. II below for a summary), but all existing schemes arguably have downsides: that of Ref. [15] requires extending the Hilbert space while that of Refs. [12, 14] requires that the perturbation is “locally reversible” and is mostly relevant to situations that exhibit a mixed anomaly between strong and weak symmetries, rather than anomalies of strong symmetries themselves that are the most relevant to defining quantum topological order [3, 9, 10]. While the definition of dressed operators from Refs. [12, 14] still applies to these, it is unclear whether they inherit the algebraic properties (particularly: anomalies) of the ‘bare’ symmetry operators, which are important in defining both the statistics of emergent excitations and the algebra of logical operators. A similar qualification applies to [16]. The applications to finite temperature phases in higher dimensions are also less explored.

In this work, we complement these existing results by showing that, under fairly general assumptions, one can explicitly construct emergent higher-form symmetries that, while related to the bare symmetries through quantum channels, nevertheless remain unitary operators on the entire Hilbert space and retain their key algebraic properties, along with their spatial locality. As in Refs. [12, 14], our definition of emergent strong symmetry generators is based on the idea of recovery channels (also referred to as decoders) that map states onto a specified code subspace that is defined by the original bare symmetry generators. In particular, we focus on cases when the symmetries are generated by a set of local commuting Pauli stabilizers, which also define a stabilizer code. Our key novel observation is that if the decoder in question has an appropriate property, namely that it is deterministic, then the resulting dressed symmetries have a particularly simple structure: they are obtained from the bare symmetries by multiplying (not conjugating!) them with a unitary that commutes with all symmetry operators by construction.

We show through explicit examples how to construct decoders that are deterministic in the appropriate sense while also being quasi-local, i.e. given by circuits of channels where the size of individual gates and the number of layers are both polylogarithmic in system size (although, as we discuss below, a weaker notion of emergent symmetry goes through even with circuits of constant, rather than polylogarithmic, locality). Whenever both properties hold simultaneously, the notion of anomalies, along with symmetry defects and their braiding/exchange statistics remain well defined and coincide with those of the bare symmetries. This leads us to propose a cleaner symmetry-based definition of topological order in mixed systems akin to the existing definitions of ground state topological phases.

Quasi-local decoders and emergent symmetries.

Our starting point is a mixed quantum state ρ\rho which we take to describe some fixed-point limit of a particular mixed state phase of matter. In particular, we assume that it has a set of exact strong higher-form symmetries Wγαρ=ρW^\alpha_\gamma\rho= \rho. Here γ\gamma labels a pp-dimensional submanifold (making WγαW^\alpha_\gamma a (d−p)(d-p)-form symmetry; e.g. k=1k=1 in the case of anyon loops) and α\alpha labels different symmetry types (e.g., different anyon species). Given such a state, one can define the type of topological order by examining the anomalies of the strong higher form symmetries, which are encoded in their algebraic relations, a discussed in [9, 10].

Our goal is to extend this discussion to states that are close to ρ\rho in a way we now make more precise. Consider a “corrupted” state ρ~\tilde{\rho}. An example to have in mind would be a state obtained by subjecting a 2D toric code state to sufficiently weak local noise. Another would be the thermal Gibbs state of the 4D toric code at sufficiently low temperatures. Common to these examples is the existence of a recovery channel R\mathcal{R} that undoes the effects of noise and takes ρ~\tilde{\rho} to an appropriate state of the noiseless toric code, R[ρ~]=ρ\mathcal{R}[\tilde{\rho}]=\rho. It is known [2, 14, 15] (see also Refs. [17, 18, 19] for related discussions) that we can use the adjoint of this recovery map to define “dressed” operators W~γα≡R†[Wγα]\widetilde{W}^\alpha_\gamma\equiv\mathcal{R}^{\dagger}[W^\alpha_\gamma]. These generate a strong symmetry of ρ~\tilde{\rho}, which follows [12] from

⟨W~γα∣ρ~⟩=⟨Wγα∣R[ρ~]⟩=⟨Wγα∣ρ⟩=Tr⁡(Wγαρ)=1,(1)\langle\widetilde{W}^\alpha_\gamma|\tilde{\rho}\rangle= \langle W^\alpha_\gamma|\mathcal{R}[\tilde{\rho}]\rangle= \langle W^\alpha_\gamma|\rho\rangle= \operatorname{Tr}(W^\alpha_\gamma\rho) = 1, \tag*{(1)}

where ⟨⋅∣⋅⟩\langle\cdot|\cdot\rangle denotes the Hilbert-Schmidt inner product between operators.

To make this definition of dressed symmetry operators physically meaningful, R\mathcal{R} has to satisfy certain requirements. One such requirement, discussed in [14] is locality. R†\mathcal{R}^{\dagger} should be locality-preserving, such that W~γα\widetilde{W}^\alpha_\gamma is still meaningfully localized near the submanifold labeled by γ\gamma, at least when this submanifold is sufficiently large. As we will discuss below, in the examples mentioned (as well as many others), R\mathcal{R} can be made to act locally on a scale that is logarithmic in the linear size of the system LL, assuming we are willing to tolerate a small error in the relation R[ρ~]=ρ\mathcal{R}[\tilde{\rho}]=\rho that vanishes in the thermodynamic limit. If we only want Eq. (1) to hold up to a small O(1)O(1) error ϵ\epsilon, then we will be able to make R\mathcal{R} local on a scale that depends on ϵ\epsilon but is independent of LL. In either case, the geometry of the dressed operators will be maintained on sufficiently large scales.

However, there remains a question about the algebraic properties of these dressed symmetries. In particular, whether they preserve the anomaly structure of the original higher-form symmetries, or, in the language of error correction, whether the symmetry operators associated to non-contractible surfaces satisfy the same logical algebra. In particular, since the dressed operators are obtained by applying a (dual) quantum channel, a priori they have no reason to be unitary operators. Nevertheless, we will now argue that in many cases there are natural choices of R\mathcal{R} for which in fact they do remain unitary and also maintain their algebraic relations. It should be mentioned that although W~γα\widetilde{W}^\alpha_\gamma is not in general unitary, one can show that an extended state ρ~⊗∣0⟩⟨0∣anc\tilde{\rho}\otimes|0\rangle\langle0|_{\mathrm{anc}} possesses a dressed versions of the original symmetry [12, 14, 15] with the correct algebraic properties. In this work, we will attack the problem of trying to define appropriate dressed symmetry operators, with the original algebraic relations, acting on the original Hilbert space.

We will now show how this can be done in a large family of cases of interest. Let us assume that the uncorrupted state ρ\rho is defined in terms of a stabilizer code so that we can partition the Hilbert space into subspaces labeled by the syndrome ss, i.e. the set of stabilizers being violated. To each subspace, we assign a projector Πs\Pi_s such that ∑sΠs=1\sum_s \Pi_s = 1. The bare symmetry operator WγαW^\alpha_\gamma is part of the stabilizer group and is thus a Pauli operator that is block diagonal w.r.t. this decomposition: Wγα=∑sΠsWγαΠsW^\alpha_\gamma= \sum_s \Pi_s W^\alpha_\gamma\Pi_s. Let us further assume that the Kraus operators of the recovery channel are of the form Ks=VsΠsK_s = V_s\Pi_s where VsV_s is a Pauli operator that corrects states within the sector Πs\Pi_s. As VsV_s and WγαW^\alpha_\gamma are both Pauli operators, they obey WγαVs=VsWγαeiθγ,sαW^\alpha_\gamma V_s = V_s W^\alpha_\gamma e^{i\theta^\alpha_{\gamma,s}}, (for a qubit stabilizer code θγ,sa=0,π\theta^{a}_{\gamma,s}=0,\pi encodes as the two Paulis can only either commute or anti-commute but we also allow more general qudit codes, which can be used to realize any 2D Abelian topological order that admits a gapped boundary [20]. From this we find

W~γa=∑sΠsVs†WγaVsΠs=∑seiθγ,saΠsWγaΠs≡UγaWγa,(2)\widetilde{W}^{a}_{\gamma}=\sum_{s}\Pi_{s}V_{s}^{\dagger}W^{a}_{\gamma}V_{s}\Pi_{s}=\sum_{s}e^{i\theta^{a}_{\gamma,s}}\Pi_{s}W^{a}_{\gamma}\Pi_{s}\equiv U^{a}_{\gamma}W^{a}_{\gamma}, \tag*{(2)}

where Uγa=∑seiθγ,saΠsU^{a}_{\gamma}=\sum_{s}e^{i\theta^{a}_{\gamma,s}}\Pi_{s} is itself a unitary operator. W~γa\widetilde{W}^{a}_{\gamma} is thus also unitary. Moreover, since UγaU^{a}_{\gamma} is proportional to the identity within each ss-block, it commutes with any of the symmetry operators and thus preserves their algebraic structure. In particular, we may view UγaU^{a}_{\gamma} is a unitary function of stabilisers operators, and as such commutes with any bare symmetry operator Wγ′a′W^{a'}_{\gamma'}. It follows immediately that dressed and bare symmetry operators have the same commutation relations [Wγa:Wγ′a′]=[W~γa:W~γ′a′][W^{a}_{\gamma}:W^{a'}_{\gamma'}]=[\widetilde{W}^{a}_{\gamma}:\widetilde{W}^{a'}_{\gamma'}], where [A:B]≡A−1B−1AB[A:B]\equiv A^{-1}B^{-1}AB.

To calculate anomalies, which encode braiding and self statistics, one would also want to consider restricted versions of symmetry operators, which have boundaries where stabilizers are violated (e.g. open line operators creating pairs of anyons). The situation here is slightly more complicated. One can define UγaU^{a}_{\gamma} in a similar manner, by keeping track of the phases that arise when commuting WγaW^{a}_{\gamma} (with γ\gamma now being an open submanifold, with boundaries) through VsV_{s}. However, the operators UγaU^{a}_{\gamma} and Uγ′bU^{b}_{\gamma'} need no longer commute once at least one of γ,γ′\gamma,\gamma' have boundaries. Nevertheless, as we argue in App. § 1.8, the operators W~γa\widetilde{W}^{a}_{\gamma} maintain the anomalies of their undressed counterparts as long as the distances between the excitations remain larger than the scale on which the decoder R\mathcal{R} acts. This ensures that the phases θγ,sa\theta^{a}_{\gamma,s} are unaffected by the introduction of these new excitations and therefore drop out of the calculation of the anomaly. Thus, the dressed symmetries have the same anomalies as the bare ones.

One interpretation of this result is as follows. Consider dividing the Hilbert space into different logical sectors based on the eigenvalues of some complete set of logical “Z” operators. Within each sector, there is one logical (ground) state as well as many states containing syndromes. We can label a generic basis state as ∣z,s⟩|z,s\rangle, where zz denotes the logical sector and for each syndrome ss we can define a Pauli operator VsV_{s} such that ∣z,s⟩=Vs∣z,0⟩|z,s\rangle=V_{s}|z,0\rangle. Let now WγxW^{x}_{\gamma} be a logical “X”-type operator, defined for some non-contractible γ\gamma, that maps code states from one sector to another: Wγx∣z,0⟩=∣z′,0⟩W^{x}_{\gamma}|z,0\rangle=|z',0\rangle where z′z' is determined by zz and WγxW^{x}_{\gamma}. Applying the same WγxW^{x}_{\gamma} to a generic basis state we get Wγx∣z,s⟩=eiθγ,sx∣z′,s⟩W^{x}_{\gamma}|z,s\rangle=e^{i\theta^{x}_{\gamma,s}}|z',s\rangle, picking up an additional phase. What the above definition of W~γx\widetilde{W}^{x}_{\gamma} achieves is to undo this phase factor so that W~γx∣z,s⟩=∣z′,s⟩\widetilde{W}^{x}_{\gamma}|z,s\rangle=|z',s\rangle. This new, dressed logical operator thus acts the same way as WγxW^{x}_{\gamma} on the entire Hilbert space, irrespective of the syndromes. Once these dressed operators are defined for non-contractible sub-manifolds, we can combine them to form operators associated to contractible ones.

The key ingredient in this discussion was that the decoder R\mathcal{R} acts deterministically, meaning that for each syndrome ss we associate a unique correction VsV_{s}. We could instead have a decoder that chooses randomly between multiple different correction operators, all of which get rid of the same syndrome. In this case, we would no longer end up with a unique phase factor θγ,s\theta_{\gamma,s} for a given syndrome and the resulting effect will not generally be unitary.

The above construction yields a dressed logical that has the correct effect in the whole Hilbert space. This cannot be achieved in a local way due to syndromes that are created by highly non-local errors VsV_{s}. However, we do not in fact need to get the phase factors θγ,s\theta_{\gamma,s} correctly for all syndromes, merely for ones that actually have a significant probability of occurring in ρ~\widetilde{\rho}. We can thus construct an approximate decoder R\mathcal{R} that still manages to approximately clean up the errors in ρ~\widetilde{\rho} (and thus restore the exact symmetries) while R\mathcal{R} is both locality-preserving and deterministic in the above sense. Below, we will discuss how to construct such decoders in a few paradigmatic cases.

The degree of locality we can achieve depends on the precise condition we require from our approximate recovery map. Conceptually, the simplest condition is to ask that R[ρ~]\mathcal{R}[\widetilde{\rho}] and ρ\rho are close to each other in trace norm. This means that R\mathcal{R} cleans up all syndromes in ρ~\widetilde{\rho} with high probability and has the advantage that such a recovery can be used to dress any symmetry operators, including those whose size scales with the system size LL, which is needed to have a non-trivial logical action. However, as the examples below will demonstrate, this typically requires that R\mathcal{R} is only quasi-local, in the sense that when we write it as a finite-depth circuit it contains gates whose support is of size polylog⁡(L)\operatorname{polylog}(L). In this case one can make ∥R[ρ~]−ρ∥1\lVert\mathcal{R}[\widetilde{\rho}]-\rho\rVert_{1} vanish faster than any polynomial of 1/L1/L, so that the symmetries become exact in the thermodynamic limit; however, these symmetries are only well-defined on scales larger than the light-cone of R†\mathcal{R}^{\dagger} which is itself diverging (polylogarithmically) with LL in this case. This is due to the fact that in a system of linear size LL, one is likely to find a syndrome of size O(log⁡L)O(\log L) somewhere in the system.

On the other hand, to apply the definition of topological order in Refs. [9, 10], which involves strong higher-form symmetries and their anomalies, it is sufficient to consider symmetry operators acting on sub-manifolds γ\gamma whose size remains finite in the thermodynamic limit (but is much larger than the microscopic correlation length). Considering such a γ\gamma, to define the dressed symmetry operator W~γa=R†[Wγa]\widetilde{W}^{a}_{\gamma}=\mathcal{R}^{\dagger}[W^{a}_{\gamma}], it is sufficient for the decoder R\mathcal{R} to clean up those syndromes that would spoil the bare symmetry operator WγaW^{a}_{\gamma}. As γ\gamma has finite size, it is unlikely to encounter any of the rare syndromes that cannot be corrected locally, by pairing syndromes within a finite distance. In particular, we can ensure tr⁡(R†[W~γ]ρ~)=1+ϵ\operatorname{tr}(\mathcal{R}^{\dagger}[\widetilde{W}_{\gamma}]\widetilde{\rho})=1+\epsilon with a decoder R\mathcal{R} that has circuit depth O(polylog⁡(1/ϵ,∣γ∣))O(\operatorname{polylog}(1/\epsilon,|\gamma|)) that is independent of LL for operators with a finite support ∣γ∣|\gamma|. Below, for simplicity, we stick to the case with O(polylog⁡(L))O(\operatorname{polylog}(L)) depths that ensure vanishing errors in the thermodynamic limit, but adapting our proofs to the finite-depth case should be straightforward.

The decohered toric code model

As a paradigmatic example, let us consider states ρ~\widetilde{\rho} obtained by subjecting a pure ground state of the 2D toric code to a weak, finite-depth error channel, such as the ones discussed recently in [21, 22, 23, 24, 25]. Our goal is to exhibit a decoder that inverts the effects of the noise while being both deterministic (assigning a unique correction Pauli for each syndrome) and quasi-local. We do so by adapting the renormalization group decoder of [26].

A key observation behind the RG decoder is a decomposition of syndromes into connected clusters. We call a set kk-connected if any bipartitioning of it has the resulting two sets within distance at most kk of one another. In step pp of the RG decoder algorithm, one identifies 2p2^{p}-connected clusters of the syndromes and, if they are are “anyon neutral” then corrects them locally. As we show in App. B, for sufficiently small noise (e.g., ϵ<10−8\epsilon<10^{-8} is a very conservative bound), this procedure terminates before level p=p∗=1+10log⁡log⁡Lp=p_{*}=1+10\log\log L, and all errors are removed with probability 1−O(L−∞)1-O(L^{-\infty}). Another result of this argument is that, with the same high probability, all of the removed 2p2^{p}-connected clusters of errors fit inside boxes with maximum side-length ℓ∗=2p∗=21+10log⁡log⁡L=O(polylog⁡(L))\ell_{*}=2^{p_{*}}=2^{1+10\log\log L}=O(\operatorname{polylog}(L)).

To make the RG decoder channel local, we partition the 2D lattice into square boxes of size ℓp=6ℓ∗\ell_{p}=6\ell_{*}. We consider 4 such partitions, with the boxes shifted relative to each other as in Figure 1; we call these four partitions Λ0,1,2,3\Lambda_{0,1,2,3} respectively. We now break down each step of the RG decoder into 4 sub-steps. In sub-step 0 we consider only the 2p2^{p}-connected clusters that are enclosed entirely within one of the boxes within Λ0\Lambda_{0} and have a distance >2p>2^{p} from the boundary of the enclosing box. For such clusters, we correct them if they are anyon neutral and leave them alone otherwise. We then repeat, with the same pp, for 2p2^{p}-connected clusters within the boxes corresponding to Λ1\Lambda_{1}, Λ2\Lambda_{2} and finally Λ3\Lambda_{3}. We then iterate pp until p=p∗p=p_{*}. All clusters that would have been corrected by the original RG decoder by the p∗p_{*}-th step are corrected by the new procedure; the key point, elaborated in App. B, is that for any p≤p∗p\leq p_{*}, all 2p2^{p}-connected clusters lie comfortably inside a box of length ℓ∗\ell_{*}. This box in turn lies entirely within one of the length 6ℓ∗6\ell_{*} boxes in one of the partitions Λ0,1,2,3\Lambda_{0,1,2,3}, and is separated at least by 2p2^{p} (indeed by ℓ∗\ell_{*}) from the boundary, so will be removed by our modified algorithm.

Defect cleaning algorithm at scale $2^p$

Figure 1. Defect cleaning algorithm at scale 2p2^p. The algorithm proceeds to find 2p2^p-connected clusters of defects contained entirely within the dashed boxes, at least a distance 2p2^p from the border of the boxes. At each step (a)-(d) the positions of the boxes are shifted by half their widths in the Cartesian directions.

Finite temperature SSB phases

While the above discussion focused on states with topological order, the techniques generalize to more traditional Landau-Ginzburg orders and furnish a method for defining dressed unitary order parameters away from idealized fixed points. As an example, consider the Ising model in D=2D=2. The ground states of this model HIsing=−∑⟨ij⟩ZiZjH_{\mathrm{Ising}}=-\sum_{\langle ij\rangle}Z_iZ_j are stabilizer states defined by ZiZj=IZ_iZ_j=I and ∏jXj=±I\prod_jX_j=\pm I. In other words, the model has a Z2\mathbb{Z}_2 order parameter (Zi2=IZ_i^2=I) which exhibits long-range order and is Ising odd—i.e., has an anomaly with the global Ising symmetry. We will show that the finite temperature Ising model, below the ordering transition, has dressed versions of these ZZ operators, exhibiting long-range order and the same algebraic conditions. Concretely, below the ordering transition temperature β>βc\beta>\beta_c the state ρβ∝e−βHIsing\rho_\beta\propto e^{-\beta H_{\mathrm{Ising}}} has dressed order parameters obeying Z~iZ~jρβ=ρβ+O(L−∞)\widetilde{Z}_i\widetilde{Z}_j\rho_\beta=\rho_\beta+O(L^{-\infty}) provided d(i,j)≥O(log⁡(L)1+ϵ)d(i,j)\geq O(\log(L)^{1+\epsilon}). The dressed order parameters have the additional desirable properties of 1) unitarity, Z~i2=I\widetilde{Z}_i^2=I; 2) Ising odd-ness {∏jXj,Z~i}=0\{\prod_jX_j,\widetilde{Z}_i\}=0 and by construction, 3) (quasi)-locality—they are smeared out over a diameter O(log⁡(L)1+ϵ)O(\log(L)^{1+\epsilon}) region.

To show this, it suffices to construct a deterministic decoder analogous to those discussed for topological phases. In the low-temperature ordered phase, a domain wall in the Pauli ZZ basis has a probability that is exponentially suppressed in length. The probability of having a domain wall of length ℓ\ell somewhere in the system is upper bounded by LDe−αℓL^De^{-\alpha\ell} where α>0\alpha>0 vanishes as one approaches the critical temperature from below, and DD is spatial dimension. As a result, the probability of having a domain wall larger than log⁡(L)1+ϵ\log(L)^{1+\epsilon} is O(L−∞)O(L^{-\infty}) for any ϵ>0\epsilon>0. Thus any decoder which removes domain walls up to size log⁡(L)1+ϵ\log(L)^{1+\epsilon} will succeed in decoding the state to a zero temperature/domain-wall free configuration with high probability, i.e., R(ρβ)=ρ∞+O(L−∞)\mathcal{R}(\rho_\beta)=\rho_\infty+O(L^{-\infty}).

Figure 2 describes a box decoder which achieves this. The decoder is similar in spirit to that discussed in [27]. It splits the system into boxes of size ℓB=log⁡(L)1+ϵ\ell_B=\log(L)^{1+\epsilon}. At each step, the grid of boxes is translated as in the figure. The domain walls entirely contained within the boxes are deterministically removed at each stage by flipping the corresponding domain. Once completed, the only remaining domain walls are necessarily at least 12log⁡(L)1+ϵ\frac{1}{2}\log(L)^{1+\epsilon} in extent; but the probability of there being such a loop scales as LDe−O(log⁡(L)1+ϵ)=O(L−∞)L^De^{-O(\log(L)^{1+\epsilon})}=O(L^{-\infty}) in the low temperature phase. It is straightforward to then argue that the resulting decoder recovers the zero temperature state with high fidelity R(ρβ)=ρ∞+O(L−∞)\mathcal{R}(\rho_\beta)=\rho_\infty+O(L^{-\infty}). Moreover, the decoder is deterministic because at each stage the spin flip to be applies is fully determined by the syndrome (i.e., domain wall) configuration.

Loop cleaning algorithm for the finite temperature Ising model

Figure 2. Loop cleaning algorithm for the finite temperature Ising model. The algorithm proceeds examines a grid with boxes of size [log⁡(L)]1+ϵ[\log(L)]^{1+\epsilon}. Whenever domains are entirely contained in a box, the domain is removed (red). The positions of the boxes are shifted by half a box length in every possible direction, and the process repeated. (There are 2d2^d rounds, where dd is the spatial dimension). By the end of the process only loops of length greater than 12[log⁡(L)]1+ϵ\frac{1}{2}[\log(L)]^{1+\epsilon} (purple) remain.

By the same arguments as above R†(ZiZj)ρβ=ρβ+O(L−∞)\mathcal{R}^{\dagger}(Z_iZ_j)\rho_\beta=\rho_\beta+O(L^{-\infty}). Moreover, provided dist⁡(i,j)>O(log⁡(L)1+ϵ)\operatorname{dist}(i,j)>O(\log(L)^{1+\epsilon}) we have R†(ZiZj)=Z~iZ~j\mathcal{R}^{\dagger}(Z_iZ_j)=\widetilde{Z}_i\widetilde{Z}_j where Z~i=R†(Zi)\widetilde{Z}_i=\mathcal{R}^{\dagger}(Z_i). Putting these equations together

Z~iZ~jρβ=ρβ+O(L−∞).(3)\tilde{Z}_i\tilde{Z}_j\rho_\beta=\rho_\beta+O(L^{-\infty}). \tag*{(3)}

As above, we can use the deterministic property of the decoder to show that Z~i=(−1)f({ZjZk})Zi\tilde{Z}_i=(-1)^{f(\{Z_jZ_k\})}Z_i for an integer function ff of stabilizers ZiZjZ_iZ_j. The phase prefactor (−1)f(-1)^f is the parity number of domain walls that cross site ii during the decoding process. This dressed operator has the required properties of 1) unitarity Z~i2=I\tilde{Z}_i^2=I, 2) Ising odd-ness {∏jXj,Z~i}=0\{\prod_jX_j,\tilde{Z}_i\}=0 and 3) satisfying Eq. (3).

Higher dimensions and finite temperature topological order

Our results naturally apply to topologically ordered states in finite-temperature equilibrium systems. While such systems are not known to exist in two dimensions, they do exist in higher dimensions. A well-known example if the four-dimensional “loop-loop” toric code [28].

The first example of finite-temperature topological order in three dimensions was constructed in [27].

The key ingredient which makes these various example stable at finite temperature is that the relevant excitations are extended loop-like objects, similar to the domain walls of the 2D Ising model discussed above, rather than the point-like anyons of 2D topological order. This makes a quasi-local correction of syndromes much more straightforward, as was already pointed out in [27]. In particular, we can provide such a decoder using a slight generalization of the construction for the 2D Ising model above. We cut the system into boxes, performing error correction in each box separately. Due to the loop-like nature of excitations, we need not resort to an RG decoder here (although such a scheme should also work); instead, we can simply consider any loop that is entirely contained within a given box, which can be corrected by a Pauli operator also contained within it. We repeat this for an appropriate number of shifted boxes (consisting of 2D2^D sub-steps in total). After this, all loops of linear size smaller than ℓ\ell have been corrected, without the need of any further iteration. A Peierls argument, crucial for finite-temperature stability, ensures that the probability of finding a loop of size ≥ℓ\ge\ell decays as LDe−cℓL^{D}e^{-c\ell}, such that ℓ=O(log⁡aL)\ell= O(\log^{a} L) with a>1a > 1 suffices for almost-perfect recovery.

In the case of the 3D fermionic toric code, whose central feature is the fermionic self-statistics of point-like emergent excitations, one also wants to use this prescription to define the appropriate dressed line operators that create these local fermions. As discussed in Sec. II, the (quasi-)locality of the decoders ensures that such self-statistics are also inherited from the ‘bare’ symmetries, as we argue in App. § 1.8. As we also show in App. § 1.8, however, for the particular case of the fermionic toric code, there is a more straightforward proof of fermionic self-statistics that does not rely on the aforementioned locality argument.

In this section, and in Sec. IV we focused on low temperature thermal states of stabilizer models. However, one can show that in many cases, the Peierls argument survives the addition of small perturbations (with the appropriate 0-form symmetry in the Ising case) to the Hamiltonian, so that loop excitations (defined in terms of the unperturbed Hamiltonian) remain (polylogarithmically) small [29, 30, 31]. Our procedure then defines emergent symmetries for generic states within the same finite-temperature phase sufficiently close to the fixed point. Moreover, similar arguments could be made for states subject to sufficiently weak local noise (again with a symmetry restriction in the Ising case).

Outlook

In this work, we focused on systems that are in some sense sufficiently close to some commuting stabilizer model. Namely, those that can be corrected back to the stabilizer fixed point using some (quasi-local) deterministic recovery map R\mathcal{R}. We then showed that the adjoint of the recovery map dresses symmetries in a way that maintains their relevant properties, i.e. both their locality and their symmetry anomalies, the latter of which encode the braiding and self-statistics of excitations as well as the algebra of logical operators. We used explicit constructions to show that an appropriate R\mathcal{R} can be constructed in many cases of interest, including the 2D toric code subject to weak noise and various finite temperatures phases of matter based on (perturbations of) stabilizer Hamiltonians.

For a given recovery map R\mathcal{R}, the procedure will break down once we go beyond its error threshold, which will likely happen, at least in some cases, before we encounter a genuine intrinsic phase transition (e.g. either a finite-temperature thermodynamic transition or a divergence of the Markov length). This is different from the situation in zero temperature phases, where quasi-adiabatic continuation ensures the existence of dressed symmetry operators everywhere within a gapped phase. It may be possible to construct such local unitary dressed symmetry operators also within the entire mixed state phase by combining a series of smaller recovery maps along the lines of [12, 14]. Proving, or disproving this conjecture is an interesting open problem.

Another question concerns finite temperature phases perturbed away from the stabilizer fixed point. When quasi-local deterministic decoders can be applied to such models [29, 30, 31], the existence of unitary dressed symmetries (particularly, of logical operators) should be reflected in the structure of the eigenstates of these Hamiltonians, including the states finite energy-density on which the finite-temperature Gibbs state is concentrated. Working out these implications for highly excited eigenstates is an interesting problem.

Lastly, in this work, we used stabilizer models as our starting point, which rules out e.g. topological orders with non-Abelian anyons. It is interesting to ask to what extent our results generalize to more general cases, including non-Abelian string-net models. These can still be described using commuting projectors (at least within the ground space) but not generally stabilizers. As a result, the operators that pair up non-abelian syndromes do not have the simple algebraic relations leading to Eq. (2); in particular, the second equality in that expression fails to generalize. It is likely that new ideas are required to extend our results to non-abelian models, if it is indeed possible to do so.

Acknowledgments

We thank Meng Cheng, Carolyn Zhang and Samuel Garratt for helpful discussions. S.G. acknowledges support from the Co-design Center for Quantum Advantage (C2QA) under contract number DE-SC0012704. C.K. is supported by a UKRI FLF MR/Z000297/1. T.R. is supported by the HUN-REN Welcome Home and Foreign Researcher Recruitment Programme 2023, and by the Supported Research Groups Programme, HUN-REN-BME-BCE Quantum Technology Research Group (TKCS-2024/34). This research was initiated at the Kavli Institute for Theoretical Physics (KITP), supported in part by grant NSF PHY-2309135. ChatGPT-6 Astra was used to improve the readability of the text.

References

Braiding and self-statistics

We now discuss how to define anomalies locally, using dressed versions of operators that create excitations and argue that the anomalies of the bare operators are preserved by the dressing, provided that the excitations remain further apart than the length scale associated with the decoder R\mathcal{R}. We focus on the case of point-like particles for simplicity.

Consider the operator WγaW_{\gamma}^{a} in the stabilizer model, where γ\gamma is an open line that creates a particle aa and its anti-particle aˉ\bar{a} at its two endpoints. Similarly as in the main text, we define the dressed creation operator as W~γa=WγaUγa\widetilde{W}_{\gamma}^{a}=W_{\gamma}^{a}U_{\gamma}^{a} with Uγa=∑seiθγ,sΠsU_{\gamma}^{a}=\sum_{s}e^{i\theta_{\gamma,s}}\Pi_{s}, where θγ,s\theta_{\gamma,s} is the phase picked up when commuting WγaW_{\gamma}^{a} through VsV_{s}. Note that now W~γa\widetilde{W}_{\gamma}^{a} itself is block off-diagonal in the syndrome sectors Πs\Pi_{s} and therefore does not commute with UγaU_{\gamma}^{a}; whether we put UγaU_{\gamma}^{a} to the left or right of WγaW_{\gamma}^{a} is an arbitrary choice. On the other hand, the dressing UγbU_{\gamma}^{b} does commute with WγbW_{\gamma}^{b} for a closed loop γ′\gamma' for any particle type bb, a fact we will use below. We now want to argue that the presence of the dressing UγaU_{\gamma}^{a} does not change either the braiding or the self-statistics of the particles $a

Braiding statistics

Let us first consider braiding statistics between particles of type aa and bb, which we can extract by considering the commutation between an open string W~γa\widetilde{W}^a_\gamma that creates a particle aa at some particular location and a closed loop W~γ′b\widetilde{W}^b_{\gamma'} encircling it, which has aa on the inside and its anti-particle aˉ\bar{a} on the outside of the loop γ′\gamma' (see Figure 3). The braiding statistics in the original stabilizer model (corresponding to anomalies of the strong symmetries of the uncorrupted state ρ\rho) is encoded in the group commutator [Wγa:Wγ′b]=eiαab[W^a_\gamma: W^b_{\gamma'}] = e^{i\alpha_{ab}}. We now consider the group commutator of the corresponding dressed operators:

Figure 3. Figure demonstrating the braiding protocol in the text. The solid lines represent the bare operators WγaW^a_\gamma, Wγ′bW^b_{\gamma'}, while the shadings represent the dressings UγaU^a_\gamma, Uγ′bU^b_{\gamma'} due to the decoder.

W~γaW~γ′bW~γa†W~γ′b†=WγaUγaWγ′bUγ′bUγa†Wγa†Uγ′b†Wγ′b†=WγaWγ′bUγ′bWγa†Uγ′b†Wγ′b†,(4)\begin{aligned} \widetilde{W}^a_\gamma\widetilde{W}^b_{\gamma'} \widetilde{W}^{a\dagger}_\gamma\widetilde{W}^{b\dagger}_{\gamma'} &= W^a_\gamma U^a_\gamma W^b_{\gamma'} U^b_{\gamma'} U^{a\dagger}_\gamma W^{a\dagger}_\gamma U^{b\dagger}_{\gamma'} W^{b\dagger}_{\gamma'} \\ &= W^a_\gamma W^b_{\gamma'} U^b_{\gamma'} W^{a\dagger}_\gamma U^{b\dagger}_{\gamma'} W^{b\dagger}_{\gamma'}, \tag*{(4)} \end{aligned}

where we used the the fact that UγaU^a_\gamma commutes with both Wγ′bW^b_{\gamma'} and Uγ′bU^b_{\gamma'} and thus cancels with Uγa†U^{a\dagger}_\gamma.

It remains to examine the operator Uγ′bWγa†Uγ′b†U^b_{\gamma'}W^{a\dagger}_\gamma U^{b\dagger}_{\gamma'}. Imagine acting with this operator on a state with syndrome ss. The operator Wγa†W^{a\dagger}_\gamma will create a new syndrome configuration which we denote by s−∂γs-\partial\gamma. We thus have that

Wγa†=∑sΠs−∂γWγa†ΠsW^{a\dagger}_\gamma=\sum_s \Pi_{s-\partial\gamma}W^{a\dagger}_\gamma\Pi_s

and

Uγ′bWγa†Uγ′b†=∑sΠs−∂γWγa†Πsei(θγ′,s−∂γb−θγ′,sb).(5)U^b_{\gamma'}W^{a\dagger}_\gamma U^{b\dagger}_{\gamma'}=\sum_s \Pi_{s-\partial\gamma}W^{a\dagger}_\gamma\Pi_s e^{i(\theta^b_{\gamma',s-\partial\gamma}-\theta^b_{\gamma',s})}. \tag*{(5)}

To complete the argument, we assume that the decoder R\mathcal{R} has some light-cone length scale ℓ\ell which is much smaller than the distance of γ′\gamma' from the anyons created by WγaW^a_\gamma, ℓ≪dist⁡(γ′,∂γ)\ell\ll\operatorname{dist}(\gamma',\partial\gamma). In calculating the effect of R†\mathcal{R}^\dagger on Wγ′bW^b_{\gamma'} we could therefore restrict to a region Bℓ(γ′)B_\ell(\gamma') of width ℓ\ell around γ′\gamma' that does not overlap with ∂γ\partial\gamma. As such, the phase θ\theta will be insensitive to changing the syndrome configuration at ∂γ\partial\gamma and we find that

θγ′,s−∂γb=θγ′,sb, ∀s⇒Uγ′bWγa†Uγ′b†=Wγa†.(6)\theta^b_{\gamma',s-\partial\gamma}=\theta^b_{\gamma',s},\ \forall s\quad\Rightarrow\quad U^b_{\gamma'}W^{a\dagger}_\gamma U^{b\dagger}_{\gamma'}=W^{a\dagger}_\gamma. \tag*{(6)}

Plugging this back into Eq. (A1) shows that the braiding phase eiαabe^{i\alpha_{ab}} remains unchanged by the dressing of the symmetry operators.

The key point in the argument was that the decoder only moves excitations around on a scale ℓ\ell, while the distance of the new excitations added by WγaW^a_\gamma from γ′\gamma' was much larger than this scale. As a result, the phase θγ′,sb\theta^b_{\gamma',s} was unaffected. Note that in the opposite limit, if the distance was smaller than ℓ\ell, we would find that the phase would change exactly in such a way as to compensate for the original braiding statistics αab\alpha_{ab}, rendering the commutator trivial. Thus, the braiding statistics remains well defined only for loops and strings on scales much larger than ℓ\ell.

configuration at ∂γ\partial\gamma and we find that

θγ′,s−∂γb=θγ′,sb, ∀s⇒Uγ′bWγa†Uγ′b†=Wγa†.(7)\theta^b_{\gamma',s-\partial\gamma}=\theta^b_{\gamma',s},\ \forall s\quad\Rightarrow\quad U^b_{\gamma'}W^{a\dagger}_\gamma U^{b\dagger}_{\gamma'}=W^{a\dagger}_\gamma. \tag*{(7)}

Plugging this back into Eq. (A1) shows that the braiding phase eiαabe^{i\alpha_{ab}} remains unchanged by the dressing of the symmetry operators.

The key point in the argument was that the decoder only moves excitations around on a scale ℓ\ell, while the distance of the new excitations added by WγaW^a_\gamma from γ′\gamma' was much larger than this scale. As a result, the phase θγ′,sb\theta^b_{\gamma',s} was unaffected. Note that in the opposite limit, if the distance was smaller than ℓ\ell, we would find that the phase would change exactly in such a way as to compensate for the original braiding statistics αab\alpha_{ab}, rendering the commutator trivial. Thus, the braiding statistics remains well defined only for loops and strings on scales much larger than ℓ\ell.

Self-statistics

We now turn to the question of self-statistics. A standard way of defining these is through a “T-junction” move, illustrated in Figure 4 (left). This involves three open strings, γ1,2,3\gamma_{1,2,3} meeting at a point, creating an anyon of type aa there. Since all three operators are of the same type, we simplify notation and simply refer to them as W1,2,3W_{1,2,3}, respectively. We are then interested in the self-statistics encoded in the relation

T junction between three open line operators

Figure 4. The left hand of the figure shows the TT junction between three open line operators W1,2,3W_{1,2,3}. The line segments γ′,γ′′\gamma^\prime,\gamma^{\prime\prime} are used later when arguing that anomalous fermionic symmetries imply long-range entanglement App. A4. In the special case of the fermionic decohered toric code stabilizer state, we show a possible explicit form of these line operators near their junction at BB on the right of the figure.

W3W2W1W3†W2†W1†=eiαa.(8)W_3W_2W_1W_3^\dagger W_2^\dagger W_1^\dagger=e^{i\alpha_a}. \tag*{(8)}

We now want to calculate the same expression for the dressed string operators which yields

W~3W~2W~1W~3†W~2†W~1†=W3U3W2U2W1U1U3†W3†U2†W2†U1†W1†.(9)\begin{aligned} \widetilde{W}_3\widetilde{W}_2\widetilde{W}_1\widetilde{W}_3^\dagger\widetilde{W}_2^\dagger\widetilde{W}_1^\dagger &=W_3U_3W_2U_2W_1U_1U_3^\dagger W_3^\dagger U_2^\dagger W_2^\dagger U_1^\dagger W_1^\dagger. \tag*{(9)} \end{aligned}

Overall, the whole expression in both Eq. (A4) and (A5) is diagonal in syndromes, since all excitations created by the WjW_j get destroyed by their adjoints. We will therefore project down the right hand side of Eq. (A5) to find

ΠsW~3W~2W~1W~3†W~2†W~1†Πs=e−i(θγ1,s−∂γ1+θγ2,s−∂γ1−∂γ2+θγ3,s−∂γ1−∂γ2−∂γ3)×e+i(θγ1,s−∂γ1−∂γ2−∂γ3+θγ2,s−∂γ2−∂γ3+θγ3,s−∂γ3)×ΠsW3W2W1W3†W2†W1†Πs,(10)\begin{aligned} \Pi_s\widetilde{W}_3\widetilde{W}_2\widetilde{W}_1\widetilde{W}_3^\dagger\widetilde{W}_2^\dagger\widetilde{W}_1^\dagger\Pi_s &=e^{-i(\theta_{\gamma_1,s-\partial\gamma_1}+\theta_{\gamma_2,s-\partial\gamma_1-\partial\gamma_2}+\theta_{\gamma_3,s-\partial\gamma_1-\partial\gamma_2-\partial\gamma_3})}\\ &\quad\times e^{+i(\theta_{\gamma_1,s-\partial\gamma_1-\partial\gamma_2-\partial\gamma_3}+\theta_{\gamma_2,s-\partial\gamma_2-\partial\gamma_3}+\theta_{\gamma_3,s-\partial\gamma_3})}\\ &\quad\times\Pi_sW_3W_2W_1W_3^\dagger W_2^\dagger W_1^\dagger\Pi_s, \tag*{(10)} \end{aligned}

where ∂γj\partial\gamma_j stands for the excitations created by WjW_j. We want to show that the phase factor in the above expression vanishes. This is ensured by satisfying the following three equations:

θγ1,s−∂γ1=θγ1,s−∂γ1−∂γ2−∂γ3(11)\theta_{\gamma_1,s-\partial\gamma_1}=\theta_{\gamma_1,s-\partial\gamma_1-\partial\gamma_2-\partial\gamma_3} \tag*{(11)}
θγ2,s−∂γ1−∂γ2=θγ2,s−∂γ2−∂γ3(12)\theta_{\gamma_2,s-\partial\gamma_1-\partial\gamma_2}=\theta_{\gamma_2,s-\partial\gamma_2-\partial\gamma_3} \tag*{(12)}
θγ3,s−∂γ1−∂γ2−∂γ3=θγ3,s−∂γ3.(13)\theta_{\gamma_3,s-\partial\gamma_1-\partial\gamma_2-\partial\gamma_3}=\theta_{\gamma_3,s-\partial\gamma_3}. \tag*{(13)}

By assumption, all three operators W1,2,3W_{1,2,3} create the same anyon excitations at the junction. Thus, the two sides of the equations in (A7)–(A9) only differ by excitations that are all far away from location of the string γj\gamma_j. By the same locality argument as in the previous section, the phases are insensitive to adding such far away excitations and thus all three equations are satisfied. Therefore, the dressing does not affect the exchange phases

W3W2W1W3†W2†W1†=W~3W~2W~1W~3†W~2†W~1†=eiαα.(14)W_3W_2W_1W_3^\dagger W_2^\dagger W_1^\dagger= \widetilde{W}_3\widetilde{W}_2\widetilde{W}_1\widetilde{W}_3^\dagger\widetilde{W}_2^\dagger\widetilde{W}_1^\dagger= e^{i\alpha_\alpha}. \tag*{(14)}

3D fermionic toric code and 2D Fermion decoherenced toric code

It is worthwhile to discuss the paradigmatic examples of the 2D fermion-decohered toric code and the 3D fermionic toric code (at T>0T>0) as special cases of the above discussion. In these models, the emergent strong symmetry corresponds to point-like excitations with fermion self-statistics, which is crucial for the non-trivial mixed-state quantum topological order in these states. These examples have some additional structure that allows for a more direct proof, without having to rely on the above argument that was based on locality. Instead, we can argue using the fact that the syndromes in these cases fall into two distinct types and the decoder only needs to clean up one type in order to restore the relevant strong symmetry. This clean up does not touch the fermionic excitations and thus leaves their exchange statistics intact as we now argue.

In the 3D fermionic toric code (see [27] for details) the syndromes fall into two categories, corresponding to two types of stabilizers. One type are associated to the square plaquettes of the 3D lattice. These obey local constraints such they have to form closed loops (i.e., every cube of the lattice needs to have an even number of excitations around it), which we refer to as “vortex loops”. The other type of stabilizers correspond to the vertices of the lattice and these can appear as isolated, point-like quasi particles, created at the two endpoints of a string-like operator WγfW_{\gamma}^{f}. It is these point-like excitations that have a fermionic self-statistics, which can be obtained by considering T-junction (Figure 4, left). The expression in Eq. (8) evaluated for WfW^f then gives a fermionic sign eiαf=−1e^{i\alpha_f}=-1.

The observation made in [27] is that the existence of a strong symmetry under WγfW_{\gamma}^{f} is equivalent to the requirement that the system has no vortex loop excitations which can be achieved, at low temperatures, via the kind of quasi-local decoder discussed in the main text that shrinks all small vortex loops to zero. Thus, in this case, the decoding procedure yields a state that is itself a highly mixed state, with a finite density of vertex excitations, but nevertheless has the exact fermionic loop symmetry (up to the O(L−∞)O(L^{-\infty}) error of the decoder).

The importance of this for the purposes of the present discussion is that the decoding operations, VVs in this case depend on only the configuration of the vortex loops and are in fact independent of the configuration of the point-like fermion excitations. Consequently, the phases αγ,γ′f\alpha_{\gamma,\gamma'}^{f} also depend only on the vortex loop configuration. Considering now the T-junction setup, the operators WγjfW_{\gamma_j}^{f} (where γj\gamma_j are now three open strings meeting at the junction) do not change the configuration of vortex loops. Therefore all the phases in Eq. (9) are equal and therefore drop out from the equation automatically.

Likewise, we can consider the fermion decohered toric code. This is a mixed state ρfdtc∝∏p(1+Fp)\rho_{\mathrm{fdtc}}\propto\prod_p(1+F_p) with stabilisers FpF_p shown in Figure 5. The state has a fermionic 1-form symmetry: For any set of plaquettes on the lattice note that [∏p∈SFp]ρfdtc=ρfdtc[\prod_{p\in S}F_p]\rho_{\mathrm{fdtc}}=\rho_{\mathrm{fdtc}}. Also note that the product can be expressed as a line operator ∏p∈SFp=W∂Sf\prod_{p\in S}F_p=W_{\partial S}^{f}, so that ρfdtc=W∂Sfρfdtc\rho_{\mathrm{fdtc}}=W_{\partial S}^{f}\rho_{\mathrm{fdtc}} has a 1-form symmetry. This loop operator is fermionic, in the sense that when it is opened up into a line operator, it creates fermionic quasiparticles, with the associated T-junction exchange statistic; we give some explicit terminations of these line operators on the right hand side of Figure 4.

Figure 5. The FpF_p stabilisers of the fermion decohered toric code consist of a product of a standard toric code plaquette operator with an adjacent vertex operator (left). The plaquette operator is a product of XX around a contour on the direct lattice; the associated vertex operator is a product of ZZ’s around the contour on the dual lattice, obtained by shifting the original contour by vector 12(1,−1)\frac{1}{2}(1,-1). The strong-1 form symmetry generators are obtained similarly (right): A product of XX’s on a loop on the lattice multiplied by a product of ZZ’s along a 12(1,−1)\frac{1}{2}(1,-1)-shifted contour on the dual lattice.

Suppose ρfdtc\rho_{\mathrm{fdtc}} is exposed to weak noise. For sufficiently weak noise, we can use the recovery procedure discussed in the main text to define a dressed 1-form symmetry. FpF_p syndromes are measured and paired up according to, for example, the RG decoder. This gives us a prescription for unitarily dressing the fermionic line operator W~γf=WγfUγf\widetilde{W}_{\gamma}^{f}=W_{\gamma}^{f}U_{\gamma}^{f} where UγfU_{\gamma}^{f} is a function purely of FpF_p stabilisers. Analogous to the 3D FTC, in the 2D FDTC the line operators can be terminated in such a way as to commute with all the defining FF stabilisers. Operationally that means that again the phases in Eq. (9) are equal and therefore drop out from the equation automatically. Equivalently [Wγf,Uγ′f]=0[W_{\gamma}^{f},U_{\gamma'}^{f}]=0 for any pair of lines γ,γ′\gamma,\gamma'.

Proving LRE from dressed fermionic symmetry operator

Ref. [27] established that the 3D fermionic toric code remains topologically ordered over a finite temperature window T<TcT<T_c, in the sense that the density matrix is provably not expressible as an incoherent mixture of short-range entangled (SRE) states. We can use the existence of a dressed symmetry operator, along with the technology in [3], to streamline the proof given there. The same mode of proof establishes that the 2D FDTC is topologically ordered below a noise threshold; indeed, we will satisfy ourselves with proving this latter statement, as the 2D setup is slightly simpler.

We expose the 2D FDTC state to some noise ρ~=Nρfdtc\widetilde{\rho}=\mathcal{N}\rho_{\mathrm{fdtc}}. Below a noise threshold, we can use a deterministic recovery procedure (cleaning up FpF_p syndromes with the RG decoder) to obtain a unitary dressed anomalous 1-form symmetry operator W~γf=WγfUγf\widetilde{W}^{f}_{\gamma}=W^{f}_{\gamma}U^{f}_{\gamma} for ρ~\widetilde{\rho}. The RG decoder gives a UγfU^{f}_{\gamma} that is at worst polylog LL local.

The W~γf\widetilde{W}^{f}_{\gamma} have some desirable properties. If we split a path γ\gamma into two halves γ=γ1∪γ2\gamma=\gamma_{1}\cup\gamma_{2}, and are careful to terminate the line operators in such a way that the endpoints do not violate the FpF_p stabilizers (as in Figure 4, right), then we have that Wγf=Wγ1fWγ2fW^{f}_{\gamma}=W^{f}_{\gamma_{1}}W^{f}_{\gamma_{2}} holds for the bare and the dressed operators. It holds by construction for the bare operators. For the dressed operators we reason as follows. Note that Uγf=∑sΠs[Wγf:Vs]U^{f}_{\gamma}=\sum_s\Pi_s[W^{f}_{\gamma}:V_s]. It follows immediately that Uγf=Uγ1fUγ2fU^{f}_{\gamma}=U^{f}_{\gamma_{1}}U^{f}_{\gamma_{2}}. Therefore W~γf=Wγ1fWγ2fUγ1fUγ2f\widetilde{W}^{f}_{\gamma}=W^{f}_{\gamma_{1}}W^{f}_{\gamma_{2}}U^{f}_{\gamma_{1}}U^{f}_{\gamma_{2}}. But by construction the UU and WW factors commute, so that

W~γf=Wγ1fUγ1fWγ2fUγ2f=W~γ1fW~γ2f.(15)\widetilde{W}^{f}_{\gamma}=W^{f}_{\gamma_{1}}U^{f}_{\gamma_{1}}W^{f}_{\gamma_{2}}U^{f}_{\gamma_{2}}=\widetilde{W}^{f}_{\gamma_{1}}\widetilde{W}^{f}_{\gamma_{2}}. \tag*{(15)}

Now, following [3], assume contrary to expectations that

ρ~=?!∑αpα[ψα],(16)\widetilde{\rho}\mathrel{\mathop{=}^{?!}}\sum_{\alpha}p_{\alpha}[\psi_{\alpha}], \tag*{(16)}

where [ψα]=∣ψα⟩⟨ψα∣[\psi_{\alpha}]=|\psi_{\alpha}\rangle\langle\psi_{\alpha}| is a SRE pure state and pα≥0p_{\alpha}\geq0. The approximate 1-form symmetry implies

tr⁡(1+W~ABCγ′f21+W~ABDγ′′f2ρ~)=1−O(L−∞)(17)\operatorname{tr}\left(\frac{1+\widetilde{W}^{f}_{ABC\gamma^{\prime}}}{2}\frac{1+\widetilde{W}^{f}_{ABD\gamma^{\prime\prime}}}{2}\widetilde{\rho}\right)=1-O(L^{-\infty}) \tag*{(17)}

This implies that there is at least one separable state for which

W~ABCγ′f∣ψα⟩,W~ABDγ′′f∣ψα⟩=∣ψα⟩+O(L−∞).(18)\widetilde{W}^{f}_{ABC\gamma^{\prime}}|\psi_{\alpha}\rangle,\widetilde{W}^{f}_{ABD\gamma^{\prime\prime}}|\psi_{\alpha}\rangle=|\psi_{\alpha}\rangle+O(L^{-\infty}). \tag*{(18)}

This readily implies that the fermionic line operators are deformable on this state, in particular

W~ABCf∣ψα⟩=W~γ′f†∣ψα⟩+O(L−∞)(19)\widetilde{W}^{f}_{ABC}|\psi_{\alpha}\rangle=\widetilde{W}^{f\dagger}_{\gamma^{\prime}}|\psi_{\alpha}\rangle+O(L^{-\infty}) \tag*{(19)}
W~ABDf∣ψα⟩=W~γ′′f†∣ψα⟩+O(L−∞).(20)\widetilde{W}^{f}_{ABD}|\psi_{\alpha}\rangle=\widetilde{W}^{f\dagger}_{\gamma^{\prime\prime}}|\psi_{\alpha}\rangle+O(L^{-\infty}). \tag*{(20)}

Assuming that ρ~\widetilde{\rho} admits a decomposition into an incoherent mixture of SRE states Eq. (16), we have further shown that at least one of these SRE states must admit deformable dressed fermionic line operators Eq. (19). Therefore, if ρ~\widetilde{\rho} admits such a decomposition, there must be a pure SRE state, admitting a fermionic symmetry operator such that 1) fermionic line operators can be joined end-to-end Eq. (15), and 2) exhibit fermionic anticommutation at a T-junction Eq. (14), and 3) are deformable in the way expected of topological 1-form symmetries (Eq. (18)). [These three conditions hold provided the separations between AA, BB, and CC, as well as between γ′\gamma^{\prime} and BB and between γ′′\gamma^{\prime\prime} and BB, exceed the combined light-cone extent of the decoder and the finite-depth circuit used to prepare ψα\psi_{\alpha} from a product state.] However, Ref. [3] showed that SRE pure states do not admit fermionic line symmetries obeying these three properties. Hence, a pure state possessing our dressed fermionic line symmetries cannot be written as a convex sum of SRE states.

We review Bravyi-Haah’s RG decoder and their proof that it has a threshold [26]. We then show that a modified ‘local’ version of the RG decoder also has a threshold. We will need a notion of hierarchies in error sets originally due to Gács [32], but our presentation closely follows that of [26].

Let EE denote the set of locations at which errors occur. Any site u∈Eu \in E is a level-0 chunk. A level-1 chunk is a subset S⊆ES \subseteq E containing two different sites u≠wu \ne w, with diameter d(S)≤Q/2d(S) \le Q/2. Inductively, a level-nn chunk is a subset S⊆ES \subseteq E containing two disjoint level-(n−1)(n-1) chunks such that d(S)≤Qn/2d(S) \le Q^n/2.

Remark 1. Note that a level-nn chunk SS has ∣S∣=2n|S| = 2^n. Also note that any given site might participate in chunks at multiple levels in the hierarchy.

Denote by EnE_n the union of all level-nn chunks. Note that E0=EE_0 = E. Also that En+1⊆EnE_{n+1} \subseteq E_n because any level-(n+1)(n+1) chunk is made up entirely of level-nn pieces. What about level-1 chunks? Well, the only way that an error vv could fail to be in E1E_1 would be if the distance between vv and its complement in EE obey d(v,E∖{v})>Q/2d(v,E\setminus\{v\}) > Q/2. This is likely for sufficiently small error rate. So when the error rate is small, E1E_1 is empty because all the noise events are well separated.

Define Fn=En∖En+1F_n = E_n\setminus E_{n+1}. A site is in this set if it is in a level-nn chunk SS which is distance >Qn+1/2> Q^{n+1}/2 from the nearest disjoint level-nn chunk. So regard FnF_n as the set of dilute errors at level-nn. Note E=⨆n≥0FnE = \bigsqcup_{n\ge0} F_n. Recall a set is rr-connected if any bipartitioning of it has the resulting two sets within distance ≤r\le r of one another. Note we can write any set as a disjoint union of rr-connected and disjoint subsets.

Theorem 2. Let Q≥6Q \ge6. Expressing Fn=⨆αFn,αF_n = \bigsqcup_\alpha F_{n,\alpha} as a disjoint union of QnQ^n-connected sets, we have that the Fn,αF_{n,\alpha} are far from their complement in EnE_n, d(Fn,α,En∖Fn,α)>Qn+1/3d(F_{n,\alpha},E_n\setminus F_{n,\alpha}) > Q^{n+1}/3. In particular d(Fn,α,Fn,β≠α)>Qn+1/3d(F_{n,\alpha},F_{n,\beta\ne\alpha}) > Q^{n+1}/3.

Proof. Consider Fn,αF_{n,\alpha}. It suffices to show that any error w∈Enw \in E_n more than distance QnQ^n from Fn,αF_{n,\alpha} is at least distance >Qn+1/3> Q^{n+1}/3 away. Assume the contrary, i.e., that Qn<d(w,Fn,α)<Qn+1/3Q^n < d(w,F_{n,\alpha}) < Q^{n+1}/3. Consider v∈Fn,αv \in F_{n,\alpha}. We must have Qn<d(v,w)<Qn+1/3Q^n < d(v,w) < Q^{n+1}/3. As v,w∈Env,w \in E_n, they must be a part of level-nn chunks Cv,CwC_v,C_w respectively. Cv,CwC_v,C_w have diameters ≤Qn/2\le Q^n/2. So as Qn<d(v,w)Q^n < d(v,w), Cv,CwC_v,C_w must be disjoint sets. However, they cannot be too far from one another as

d(Cv∪Cw)≤d(u,v)+d(Cv)+d(Cw)≤Qn+13+Qn=Qn+1(13+1Q)≤Qn+12.\begin{aligned} d(C_v \cup C_w) &\le d(u,v) + d(C_v) + d(C_w) \\ &\le\frac{Q^{n+1}}{3} + Q^n \\ &= Q^{n+1}\left(\frac{1}{3}+\frac{1}{Q}\right) \le\frac{Q^{n+1}}{2}. \end{aligned}

The latter equality holds provided Q≥6Q \ge6. But then Cv,CwC_v,C_w are both a part of En+1E_{n+1} which contradicts our assumption.

Remark 3. Note that while d(Fn,α,En∖Fn,α)>Qn+1/3d(F_{n,\alpha},E_n\setminus F_{n,\alpha}) > Q^{n+1}/3, it need not be true replacing En→EE_n \to E. That is, there can be lower level sparse errors that are close to Fn,αF_{n,\alpha}.

Bounding error sets.

Work on the DD-dimensional Euclidean lattice. Let BnB_n be a fixed cubic box of size QnQ^n and Bn+B_n^+ a padded box of size 3Qn3Q^n. Let EE be the error set. Consider the probabilities

pn=P(Bn∩En≠∅)p_n = \mathbb{P}(B_n \cap E_n \ne\varnothing)
p~n=P(Bn+⊃level-n chunk)\widetilde{p}_n = \mathbb{P}(B_n^+ \supset\text{level-}n\text{ chunk})
qn=P(Bn+⊃2 disjoint level-(n−1) chunks)q_n = \mathbb{P}(B_n^+ \supset2\text{ disjoint level-}(n-1)\text{ chunks})
rn=P(Bn+⊃1 level-(n−1) chunk).r_n = \mathbb{P}(B_n^+ \supset1\text{ level-}(n-1)\text{ chunk}).

First note that pn≤p~np_n \le\widetilde{p}_n as whenever a chunk overlaps with BnB_n it must be contained in Bn+B_n^+. Then p~n≤qn\widetilde{p}_n \le q_n because a level-nn chunk consists of two level-(n−1)(n-1) chunks. Now qnq_n requires two disjoint level-(n−1)(n-1) chunks to be present in Bn+B_n^+. The BK theorem tells us that qn≤rn2q_n \le r_n^2.

But Bn+B_n^+ is a collection of (3Q)D(3Q)^D boxes of size Qn−1Q^{n-1}. So the union bound gives rn≤(3Q)Dpn−1r_n \le(3Q)^D p_{n-1}. Putting it all together

pn≤(3Q)2Dpn−12p_n \le(3Q)^{2D}p_{n-1}^2

Iterating gives

pn≤(3Q)(22−1)2Dpn−24p_n \le(3Q)^{(2^2-1)2D}p_{n-2}^4

and eventually

pn≤(3Q)(2n−1)2Dp02n=(3Q)−2D(ϵ(3Q)2D)2n.p_n \le(3Q)^{(2^n-1)2D}p_0^{2^n} = (3Q)^{-2D}\left(\epsilon(3Q)^{2D}\right)^{2^n}.

Thus the probability that a level-nn chunk intersects a box BnB_n decays doubly exponentially quickly. Using the union bound, this culminates in the following theorem.

Theorem 4. The probability of finding a level-nn chunk anywhere in the system is bounded above by L2(3Q)−2D(ϵ(3Q)2D)2nL^2(3Q)^{-2D}\left(\epsilon(3Q)^{2D}\right)^{2^n}.

Proving a threshold theorem for the RG decoder

The RG decoder works as follows. We are given an initial set of syndromes SS. Initialise p=0p = 0, and iterate until p=p∗p = p_*; it turns out that provided the noise is small enough, the number of RG steps p∗p_* can be chosen to be O(log⁡log⁡L)O(\log\log L) and the RG decoder will succeed with a probability exceeding 1−O(L−∞)1-O(L^{-\infty}). In each step of decoding, identify the 2p2^p-connected clusters of the syndrome SS, and bound them with minimal boxes. For any such connected clusters that are anyon neutral, apply a suitable Pauli string to the 1-neighbourhood of the corresponding bounding box to eliminate the syndromes. Update SS, and increment p→p+1p \to p+1.

Theorem 5. For Q≥18Q \ge18, if the chunk decomposition of EE terminates at the mm-th level, and Qm+1<L/2Q^{m+1} < L/2 then EE is fully corrected by the RG decoder.

Proof. Let E=⋃j=0mFjE = \bigcup_{j=0}^{m} F_j be the locations of Pauli-errors. Recall that FjF_j have QjQ^j-connected components which are at least Qj+1/3Q^{j+1}/3 separated. Label these Fj,αF_{j,\alpha}; denote the 1-neighbourhood of Fj,αF_{j,\alpha}’s minimal bounding by Bj,αB_{j,\alpha}. Then by the nature of these connected components above, we have

d(Bjα)≤Qj+2,d(B_{j\alpha}) \le Q^j+2,
d(Bjα,Bkβ)≥13Qmin⁡(j,k)+1−1.d(B_{j\alpha},B_{k\beta}) \ge\frac{1}{3}Q^{\min(j,k)+1}-1.

The first condition comes from the QjQ^j-connectedness of FjαF_{j\alpha}. The second comes from d(Fjα,Ej∖Fjα)>Qj+1/3d(F_{j\alpha},E_j\setminus F_{j\alpha}) > Q^{j+1}/3. By design, the errors on each BjαB_{j\alpha} are charge neutral; because FjαF_{j\alpha} contains an isolated collection of errors (not syndromes). Thus, all of the syndromes in any BjαB_{j\alpha} can be eliminated by acting solely on BjαB_{j\alpha}.

Note that the minimal separation between any two BB-sets is Q3−1≥5\frac{Q}{3}-1 \ge5. Recall too that all syndromes are contained inside of the union ⋃j=0,αmBj,α\bigcup_{j=0,\alpha}^{m}B_{j,\alpha}.

At p=0p=0, the algorithm finds the 1-connected clusters of syndromes. Each 1-connected component must be entirely contained inside of some BjαB_{j\alpha} because any BjαB_{j\alpha} are at least 5-separated. Get rid of the resulting 1-clusters that are charge neutral. After this correction, all BjαB_{j\alpha} must still be neutral because they started as such.

At p=1p=1, the algorithm finds 2-connected clusters of errors. Once again, as all BjαB_{j\alpha} are more than 2-separated, each such 2-connected cluster lies entirely within some BjαB_{j\alpha}.

At p=2p=2, the algorithm finds 4-connected clusters of errors. At this point, the only errors remaining in a B0αB_{0\alpha} set are part of a 4-connected neutral cluster entirely contained in said B0αB_{0\alpha}. That is because any such errors are 5-separated from errors outside of B0αB_{0\alpha}, and all errors in B0αB_{0\alpha} are within 4-sites of one another (because d(B0α)≤3d(B_{0\alpha}) \le3). The algorithm thus eliminates the B0αB_{0\alpha} sets. It also may eliminate 4-connected subsets of the Bj>0,αB_{j>0,\alpha}.

We are thus at a point p=2p=2 where all syndromes contained in BjαB_{j\alpha} with Qj+2≤2p∣p=2Q^{j+2} \le2^p\big|_{p=2} have been eliminated and all remaining BB sets are anyon neutral. We will show inductively that this relation continues to hold as we iterate p≥2p \ge2.

Suppose that all syndromes contained in BjαB_{j\alpha} with Qj+2≤2pQ^{j+2} \le2^p have been eliminated. Denote by j∗j_* the maximum such jj so that Qj∗+2≤2p<Qj∗+1+2Q^{j_*+2} \le2^p < Q^{j_*+1}+2. Iterate pp. We now find 2p+12^{p+1}-connected clusters of syndromes. These are entirely contained within remaining BB sets because for j,k≥j∗+1j,k \ge j_*+1,

d(Bjα,Bkβ)≥13Qj∗+2−1≥Q3(2p−1)−1\begin{aligned} d(B_{j\alpha},B_{k\beta}) \ge\frac{1}{3}Q^{j_*+2}-1 \\ &\ge\frac{Q}{3}(2^p-1)-1 \end{aligned}
≥2p+1+2p+1−5>2p+1.\ge2^{p+1}+2^{p+1}-5 > 2^{p+1}.

The algorithm will eliminate such a cluster if it is anyon neutral. But as the cluster is entirely contained in a BjαB_{j\alpha}, it leaves the anyon charge of BjαB_{j\alpha} unchanged. To complete the induction, we need to show that once pp becomes large enough, the Bj∗+1,αB_{j_*+1,\alpha} sets are removed. So suppose that 2p<Qj∗+1+2≤2p+12^p < Q^{j_*+1}+2 \le2^{p+1}. At stage p+1p+1, the algorithm finds 2p+12^{p+1}-connected clusters. We know that if any such cluster has support in Bj∗+1,αB_{j_*+1,\alpha} it must be entirely contained in that set from the working above. On the other hand, we know that every error in Bj∗+1,αB_{j_*+1,\alpha} is part of the cluster because d(Bj∗+1)≤Qj∗+1+2≤2p+1d(B_{j_*+1}) \le Q^{j_*+1}+2 \le2^{p+1}. We also know the cluster is anyon neutral by hypothesis. Thus every Bj∗+1,αB_{j_*+1,\alpha}, and thus all errors in the (j∗+1)(j_*+1)-th level of the chunk decomposition, are eliminated by the end of the (p+1)(p+1)-step where Qj∗+1+2≤2p+1Q^{j_*+1}+2 \le2^{p+1}. □\square

Remark 6. The intuition behind this construction is that we erase BjB_j as soon as pp obeys 2p≥Qj+2≥2p−12^p \ge Q^j+2 \ge2^{p-1}. However, the BjB_j sets are widely separated compared to their size (Qj+1Q^{j+1} vs QjQ^j), and this ensures that the 2p2^p-connected sets lie entirely in the BB’s. Any remaining errors in a BjB_j set are 2p2^p-connected and must be removed at this point because d(Bj)≤Qj+2≤2pd(B_j) \le Q^j+2 \le2^p (charge neutrality is guaranteed by construction for the BB).

Application to decoding the toric code in 2D

Subject a 2D toric code ground state to Pauli dephasing errors, with local error probability ϵ\epsilon. Take ϵ\epsilon to be small (say ϵ<12(3Q)−4\epsilon< \frac{1}{2}(3Q)^{-4}). In this case, 44 implies that the probability of finding a level-mm chunk anywhere in the system is bounded above by L2pm≤L22−2mL^2p_m \le L^2 2^{-2^m}. If we set m=2log⁡log⁡Lm=2\log\log L (logs are base 2), then this probability is bounded by L2e−log⁡(L)2=O(L−∞)L^2e^{-\log(L)^2}=O(L^{-\infty}) which decays super-polynomially in system size. Thus the chunk decomposition truncates by m=2log⁡log⁡Lm=2\log\log L with high probability 1−O(L−∞)1-O(L^{-\infty}). What are the consequences for the RG algorithm?

For the purposes of specifying a sufficiently large p∗p_*, it will be convenient to take Q=25Q=2^5 from this point onwards. Recall that the mm-th level of the chunk decomposition, FmF_m, is eliminated once Qm+2≤2pQ^m+2 \le2^p. In the present case, this is guaranteed if p=5m+1p=5m+1. This leads to the corollary:

Corollary 7. For ϵ<12(3Q)−4\epsilon< \frac{1}{2}(3Q)^{-4} and Q=25Q=2^5, the RG decoder will succeed if terminated at p∗=10log⁡log⁡L+1p_*=10\log\log L+1, with a probability exceeding 1−O(L−∞)1-O(L^{-\infty}).

We now make a local version of the RG decoder. Note that the BjαB_{j\alpha} in the chunk decomposition have maximum diameter Qm∗+2≤2p∗=2log⁡(L)10=:ℓ∗Q_m^* + 2 \le2^{p^*} = 2\log(L)^{10} =: \ell^* with high probability. Let PabP_{ab} denote the set of squares of length 2R2R whose origin (lower left corner) lies at a point of form (aR+2Rm,bR+2Rn)(aR + 2Rm, bR + 2Rn), m,n∈Zm,n \in\mathbb{Z} where a,b=0,1a,b = 0,1. We pick the square lengths to be larger than the extent of the likely BjαB_{j\alpha}, e.g., 2R=6ℓ∗2R = 6\ell^*. With this choice, any BjαB_{j\alpha} that is likely to occur will lie comfortably in at least one square in the sets PabP_{ab}, so as to be at least a distance ℓ∗\ell^* from the boundary of said square.

The algorithm proceeds as follows. We iterate on pp as in Haah-Bravyi, terminating at p∗=10log⁡log⁡L+1p_* = 10\log\log L + 1. At general level pp, we first identify all 2p2^p-connected clusters of errors in each square in P00P_{00}. But we ignore 2p2^p-connected clusters that are within distance 2p2^p of the boundary of squares in P00P_{00}. We eliminate the neutral 2p2^p-connected clusters obeying this condition as before. We repeat the same procedure for P01P_{01}, P10P_{10}, P11P_{11}. By the end of this procedure, we have clearly removed a subset of the 2p2^p-connected clusters that would have been removed in the global algorithm due to the boundary constraint. In fact we have removed all of them for the following reason. Crucially any 2p2^p-connected clusters fit into a BjαB_{j\alpha}, which will fit comfortably into one of the boxes comprising a PabP_{ab}, with the distance between BjαB_{j\alpha} and the boundary of the box being at least 2p∗2^{p^*}.

So, for example, if a neutral 2p2^p connected cluster CC is sufficiently within an element P00P_{00}, it will be dealt with identically to the global algorithm. If it is close to, or straddles the boundary of a box in P00P_{00}, it must be left alone at the first stage. However, we do know that CC will lie entirely and deeply inside—at least distance 2p∗2^{p^*} from the boundary—of a square in one of P01,10,11P_{01,10,11}. Thus CC will be removed in one of these three subsequent steps. Therefore, by then end of the level-pp round, we can be sure that CC is transformed in the same way that it would be if we were to use the global algorithm.

References

  1. [1]D. Gaiotto, A. Kapustin, N. Seiberg, and B. Willett, Generalized Global Symmetries, JHEP 02, 172, arXiv:1412.5148 [hep-th].DOI
  2. [2]J. McGreevy, Generalized symmetries in condensed mat- ter, Annual Review of Condensed Matter Physics 14, 57 (2023). Mixed-state topological order and the errorfield double formulation of decoherence-induced transitions (2023), arXiv:2301.05687 [quant-ph].
  3. [3]Z. Li, D. Lee, and B. Yoshida, How much entanglement is needed for topological codes and mixed states with anomalous symmetry?, Phys. Rev. X 15, 021090 (2025).arxiv.org/abs/2405.07970
  4. [4]S. Bravyi, M. B. Hastings, and S. Michalakis, Topological quantum order: stability under local perturbations, Journal of mathematical physics 51 (2010).DOI
  5. [5]X. Chen, Z.-C. Gu, and X.-G. Wen, Local unitary transformation, long-range quantum entanglement, wave function renormalization, and topological order, Physical Review B—Condensed Matter and Materials Physics 82, 155138 (2010).DOI
  6. [6]M. B. Hastings and X.-G. Wen, Quasidiabatic continuation of quantum states: The stability of topological ground-state degeneracy and emergent gauge invariance, Physical review b 72, 045141 (2005).DOI
  7. [7]R. Ma, J.-H. Zhang, Z. Bi, M. Cheng, and C. Wang, Topological phases with average symmetries: The decohered, the disordered, and the intrinsic, Physical Review X 15, 021062 (2025).DOI
  8. [8]L. A. Lessa, R. Ma, J.-H. Zhang, Z. Bi, M. Cheng, and C. Wang, Strong-to-weak spontaneous symmetry breaking in mixed quantum states, PRX Quantum 6, 010344 (2025).DOI
  9. [9]T. D. Ellison and M. Cheng, Toward a classification of mixed-state topological orders in two dimensions, PRX Quantum 6, 010315 (2025).DOI
  10. [10]R. Sohal and A. Prem, Noisy approach to intrinsically mixed-state topological order, PRX Quantum 6, 010313 (2025).DOI
  11. [11]V. V. Albert and L. Jiang, Symmetries and conserved quantities in Lindblad master equations, Physical Review A 89, 022118 (2014).DOI
  12. [12]L. A. Lessa, S. Sang, T.-C. Lu, T. H. Hsieh, and C. Wang, Higher-form anomaly and long-range entanglement of mixed states, arXiv preprint arXiv:2503.12792 (2025).arxiv.org/abs/2503.12792
  13. [13]P.-S. Hsin, R. Kobayashi, and A. Prem, Higher-form anomalies imply intrinsic long-range entanglement, arXiv preprint arXiv:2504.10569 (2025).arxiv.org/abs/2504.10569
  14. [14]S. Sang, L. A. Lessa, R. S. K. Mong, T. Grover, C. Wang, and T. H. Hsieh, Mixed-state phases from local reversibility, Phys. Rev. X 16, 041014 (2026).arxiv.org/abs/2507.02292
  15. [15]T. D. Ellison and M. Cheng, Toward a classification of mixed-state topological orders in two dimensions, PRX Quantum 6, 010315 (2025).DOI
  16. [16]S. W. P. Kim and M. McGinley, Mixed-state topological order and error-correction thresholds in non-abelian codes: rigorous results (2026), arXiv:2607.21706 [quant-ph].arxiv.org/abs/2607.21706
  17. [17]P. Serna, A. M. Somoza, and A. Nahum, Worldsheet patching, 1-form symmetries, and ”landau-star” phase transitions (2024), arXiv:2403.04025 [cond-mat.str-el].arxiv.org/abs/2403.04025
  18. [18]A. M. Somoza, P. Serna, and A. Nahum, Self-dual criticality in three-dimensional F2 gauge theory with matter, Phys. Rev. X 11, 041000 (2021).
  19. [19]Y.-J. Liu, W.-T. Xu, F. Pollmann, and M. Knap, Information-theoretic principle of emergent 1-form symmetries, PRX Quantum 7, 043006 (2026).DOI
  20. [20]T. D. Ellison, Y.-A. Chen, A. Dua, W. Shirley, N. Tanti-vasadakarn, and D. J. Williamson, Pauli stabilizer models of twisted quantum doubles, PRX Quantum 3, 010353 (2022).DOI
  21. [21]Y. Bao, R. Fan, A. Vishwanath, and E. Altman, Mixed-state topological order and the errorfield double formulation of decoherence-induced transitions (2023), arXiv:2301.05687 [quant-ph].DOI
  22. [22]R. Fan, Y. Bao, E. Altman, and A. Vishwanath, Diagnosis of mixed-state topological order and breakdown of quantum memory, PRX Quantum 5, 020343 (2024).DOI
  23. [23]Z. Wang, Z. Wu, and Z. Wang, Intrinsic mixed-state topological order, PRX Quantum 6, 010314 (2025).DOI
  24. [24]S. Sang, Y. Zou, and T. H. Hsieh, Mixed-state quantum phases: Renormalization and quantum error correction, Phys. Rev. X 14, 031044 (2024).arxiv.org/abs/2310.08639
  25. [25]Y.-H. Chen and T. Grover, Separability transitions in topological states induced by local decoherence, Phys. Rev. Lett. 132, 170602 (2024).DOI
  26. [26]S. Bravyi and J. Haah, Quantum self-correction in the 3d cubic code model, Phys. Rev. Lett. 111, 200501 (2013).pubmed.ncbi.nlm.nih.gov/24289671
  27. [27]S.-T. Zhou, M. Cheng, T. Rakovszky, C. von Keyserlingk, and T. D. Ellison, Finite-temperature quantum topological order in three dimensions, Phys. Rev. Lett. 135, 040406 (2025).DOI
  28. [28]E. Dennis, A. Kitaev, A. Landahl, and J. Preskill, Topological quantum memory, Journal of Mathematical Physics 43, 4452 (2002).DOI
  29. [29]D. Gamarnik, B. T. Kiani, and A. Zlokapa, Slow mixing of quantum gibbs samplers, arXiv preprint arXiv:2411.04300 (2024).arxiv.org/abs/2411.04300
  30. [30]G. Xun, W. Cao, and H. Watanabe, Local-channel stability of classical Z2 lattice gauge models, arXiv preprint arXiv:2610.27939 (2026).arxiv.org/abs/2610.27939
  31. [31]C. Yin, Y. Li, T. Rakovszky, and V. Khemani, Stability of finite temperature quantum phases of matter (2026), in preparation.
  32. [32]P. Gács, Reliable computation with cellular automata, Journal of Computer and System Sciences 32, 15 (1986).

Paper details

vlnc:W_av5bs3ww
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

ChatGPT-6 Astra

View statement
Looking for related papers…
Contents