{"id":"6cf9f63b-0d26-46fd-a0cd-b3ae496ffcbf","arxiv_id":"1908.07601","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A 3D Z2 lattice gauge theory with cube holonomy stabilizers has exponentially many ground states from its boundary, immobile fracton excitations, and entanglement entropy equal to the log of a restricted ground state degeneracy.","lead":"This paper constructs a 3D lattice model whose low-energy particles include immobile fractons and fully mobile charges, with a ground state degeneracy that grows exponentially with the square of system size. It adds a new example of exotic quantum order protected by special subsystem symmetries, so condensed matter theorists studying fracton phases will want to compare it with X-cube type models.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified: the central Z2 GSD counting and commuting-projector structure hold up under cohomological check.","rationale":"The reader's weakest_assumption was the GSD counting on the 3-ball. I found that this assumption is actually correct and can be proven by a simple cohomological argument: the cube constraints equate holonomy values on opposite faces, collapsing each straight line of faces in the x, y, and z directions into one independent binary degree of freedom. Because the 3-ball is contractible, every assignment of these face-class values is realized by some spin configuration and gauge-inequivalent configurations are in one-to-one correspondence with such assignments, giving GSD = 2^(LxLy+LxLz+LyLz) exactly. The other potentially fragile step, the operator identity used in the entanglement entropy derivation, is also correct once one uses the cube identity Z^x Z^y Z^z = 1. I therefore do not see a load-bearing flaw in the central Z2 construction. The gaps noted by the reader (missing explicit 3D subsystem symmetries, deferred non-Abelian proofs) are real limitations of the paper's presentation and scope, but they do not invalidate the main claim as stated. The verdict should remain CONDITIONAL, consistent with the reader's assessment, but the specific weak assumption identified by the reader is not the true stress point.","tokens_in":20221,"tokens_out":34091,"duration_ms":310133,"concrete_test":"Compute the stabilizer rank for a 2x2x2 open-boundary lattice with the generators X_v and Z_c^mu (two independent Z per cube after the Z^x Z^y Z^z = 1 relation); if the code dimension equals 2^(LxLy+LxLz+LyLz) = 2^12, the GSD formula is confirmed. This also verifies independence of the boundary-face classes.","verdict_should_be":"UNCHANGED","load_bearing_attack":"I checked the load-bearing GSD counting in Section III B 1 and it is sound: the condition that opposite faces of every cube have equal holonomy partitions the boundary faces into LxLy+LxLz+LyLz equivalence classes (one per straight tube in each lattice direction), and every assignment to these classes defines a valid 2-cocycle on the contractible 3-ball, hence a unique gauge orbit of spin configurations. The operator identity B_c^x B_c^y B_c^z = (1/4)(1+Z^x+Z^y+Z^z) is also correct because Z^x Z^y Z^z = 1 (each link appears in two faces of a cube); the entanglement calculation is internally consistent. The remaining gaps (3D subsystem-symmetry operators are never explicitly constructed; Section V non-Abelian claims are deferred) are real but do not undermine the Z2 central claim as stated.","agreement_with_reader":"disagree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies exactly solvable Z2 lattice gauge theory models with subsystem symmetries in two and three spatial dimensions. The 2D model, equivalent to the Xu-Moore/plaquette Ising model, is claimed to have GSD = 2^{Lx+Ly} on an Lx by Ly rectangle. The central 3D model, defined by vertex operators A_v and cube operators B_c^{(x)}, B_c^{(y)}, B_c^{(z)}, is claimed to have GSD = 2^{LxLy+LxLz+LyLz} on a 3-ball (Eq. 17), with immobile mu-flux fractons and fully mobile charges; adding local 3D toric-code plaquette operators is claimed to reduce the model to the 3D toric code. The paper also computes entanglement entropy S_A for subregion A and claims S_A = log(GSD_{\\tilde A}) for a suitably restricted model (Eqs. 33 and 45). Section V sketches generalizations to arbitrary finite groups, with several claims deferred to future work.","tokens_in":20422,"tokens_out":8190,"duration_ms":77100,"significance":"If the 3D claims are fully established, the paper provides a new example of fracton-like order in a standard Z2 lattice gauge theory, with ground-state degeneracy growing exponentially in the square of the linear size and with mobile charges coexisting with immobile fractons. The model is exactly solvable through commuting projectors, and the route from the 3D toric code to the fracton-like model via condensing non-contractible ribbon excitations is conceptually interesting. The entanglement-entropy relation S_A = log(GSD_{\\tilde A}) is a useful diagnostic and is derived from first principles in the 2D case. However, the 3D GSD counting and the 3D entanglement-entropy derivation currently rest on informal geometric arguments rather than rigorous proofs; these are the load-bearing points of the paper.","major_comments":[{"comment":"The counting leading to Eq. (17) is an assertion rather than a proof. The text assumes that every ground state is represented by a set of straight, non-bending membranes ending at opposite boundary faces, that each boundary plaquette can be independently marked with a dot, and that distinct dot configurations give gauge-inequivalent states. No argument is provided for completeness (that every solution of the stabilizer constraints is of this form) or for uniqueness (that membranes in different directions do not impose additional consistency conditions, and that no two different boundary configurations are related by vertex gauge transformations). Since Eq. (17) is the central quantitative claim of the paper, a rigorous derivation, for example by explicitly solving the stabilizer conditions or using a cohomological count of the gauge orbits, should be supplied.","section":"Section III B 1, Eq. (17)"},{"comment":"The 3D entanglement-entropy formula is not derived. After Eq. (44), the paper states Eq. (45) without computing the order |G_{\\tilde A}| of the surviving subgroup or dim(H_B). The subsequent check of Eq. (33) is also arithmetically incorrect as written: the text says 'GSD_{\\tilde A} = RxRy+RxRz+RyRz + 2(RxRy+RxRz+RyRz)+2', which adds the number of boundary gauge transformations to the area-law exponent, whereas the stated S_A = 3(RxRy+RxRz+RyRz)+2 requires GSD_{\\tilde A} = 2^{3(RxRy+RxRz+RyRz)+2}, i.e., the area exponent plus the boundary-vertex count in the exponent. This error, together with the missing counting of |G_{\\tilde A}|, prevents the reader from verifying the claimed agreement with Eq. (33) in 3D.","section":"Section IV B, Eqs. (44) and (45)"},{"comment":"The claim that only operators supported in the interior region \\tilde A survive the partial trace over B is justified only by a brief statement about gauge transformations on the boundary of A. It is not demonstrated that cube operators B_c^{(\\mu)} that cross the boundary of A always vanish under the trace, nor is the dimension of the surviving subgroup computed. A detailed derivation of |G_{\\tilde A}|, including the treatment of cube operators and vertex operators near the boundary, is necessary to support the central entanglement-entropy result.","section":"Section IV B, Eq. (43)"}],"minor_comments":[{"comment":"The statements that no local vertex operator commuting with all cube operators exists for arbitrary non-Abelian groups, and that the operators satisfy the quantum double algebra of G, are asserted without proof and with the proof deferred to future work. This section should be clearly labeled as a sketch or outlook, or the missing derivations should be included.","section":"Section V B"},{"comment":"The diagrams defining A_v, B_c^{(\\mu)}, and Z_c^{(\\mu)} appear as empty placeholders in the text; the published version must ensure that these operators are explicitly shown, since the definitions are otherwise incomplete.","section":"Equations (10)-(13) and (38)-(40)"},{"comment":"The identity B_c^{(x)} B_c^{(y)} B_c^{(z)} = (1/4)(1+Z^{(x)}+Z^{(y)}+Z^{(z)}) uses the relation Z^{(x)} Z^{(y)} Z^{(z)} = 1, which is not explicitly justified in the text; this should be stated for clarity.","section":"Equation (37)"},{"comment":"The subgroup G_A should be defined explicitly as the set of products of plaquette operators b_p whose support lies entirely in region A; this would make the computation of |G_A| transparent.","section":"Section IV A, after Eq. (29)"},{"comment":"There are several typographical errors, including 'condensations of the the 3d Toric Code model' (Section III B 1), 'an TC excitation' (Section III B 1), and 'in the follwing way' (Section V B). These should be corrected.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The central 3D model is interesting and likely correct, but the manuscript as written does not yet provide rigorous support for the 3D GSD count or the 3D entanglement-entropy formula, both of which are load-bearing for the paper's main claims. The non-Abelian section is largely a research announcement. Self-citation to the companion paper [62] is significant, but the 2D entanglement result is derived here from first principles; the 3D case, however, inherits the same counting gap that affects Eq. (17)."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Candid take: the 3D model in Eq. (14) is the real content here, and it looks correct. The 2D section is explicitly a review of Xu-Moore/plaquette Ising, so don't send this to a referee for novelty there. The new 3D part — cube operators comparing opposite plaquette holonomies, the boundary-law GSD 2^(LxLy+...) on a ball, mobile charges plus immobile fluxes, and the entropy result SA = 3(...)+2 — is a legitimate new example of fracton-like order emerging from an ordinary-looking Z2 lattice gauge theory.\n\nI checked the two load-bearing calculations myself. The GSD counting, which the paper does graphically with dots on boundary faces, is sound: the ground-state condition forces straight non-bending membranes, and each independent dot choice on the three pairs of opposite faces gives a distinct gauge orbit. A rigorous proof would be better, but I don't think it's wrong. The entropy calculation in 2D is done carefully by partial traces. The 3D entropy formula is asserted more than derived: Eq. (45) appears after a statement that |G~A| has a certain size, and the relationship to restricted GSD is one paragraph. That section needs expansion, not replacement.\n\nWhere the paper is genuinely soft: the 3D subsystem symmetries are never explicitly written down, which matters because the title and the SSET claim hang on them. The non-Abelian section V is labeled as remarks, but it still makes a substantial claim about quantum double algebras without proof; either prove it or cut it to a sentence. The discussion of GSD on a 3-torus is brief and could be misread as claiming more than it proves. None of these are fatal. The central Z2 construction and its GSD are solid enough to build on.\n\nCitation pattern: the companion paper [62] is cited for the entropy/GSD relation, and the 2D version is derived independently here, so I don't see a circularity problem. The literature review is adequate.\n\nWho this is for: anyone working on fracton models, subsystem symmetry enrichment, or exactly solvable 3D spin models. It deserves a serious referee. I'd recommend acceptance after the authors either supply a proof or a numerical check of the GSD counting and spell out the entropy derivation for Eq. (45).","headline":"A plausible and genuinely new 3D fracton-like Z2 lattice model; the central counting and entropy claims hold up on inspection, though the paper leaves some proofs as pictures and sketches.","tokens_in":20959,"tokens_out":2270,"would_cite":true,"duration_ms":553177,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper claims that a $\\mathbb{Z}_2$ lattice gauge theory with subsystem symmetries in three dimensions realizes an exactly solvable fracton-like phase, with ground-state degeneracy $2^{L_xL_y+L_xL_z+L_yL_z}$ on a 3-ball, immobile flux…","keywords":["fracton-like order","subsystem symmetries","lattice gauge theory","ground-state degeneracy","toric code","entanglement entropy","exactly solvable model","quantum double"],"falsifier":"Take a small 3-ball lattice, say $L_x=L_y=L_z=2$, and directly enumerate all states satisfying every vertex and cube projector in the Hamiltonian. If the ground-state dimension is not $2^{12}=4096$ as Eq. (17) predicts, or if two different boundary dot configurations produce the same logical state, then the straight-membrane counting assumption fails.","tokens_in":20022,"feed_emoji":"⚛️","tokens_out":10003,"duration_ms":93794,"temperature":0.7,"pith_summary":"The paper introduces exactly solvable $\\mathbb{Z}_2$ lattice gauge theories whose Hamiltonians include vertex gauge projectors and cube holonomy comparisons, and argues these models display fracton-like order: some excitations are immobile fractons living at membrane corners, while others are fully mobile charges. Its central quantitative claim is Eq. (17): on a 3-ball of side lengths $L_x,L_y,L_z$, the 3D model has ground-state degeneracy $2^{L_xL_y+L_xL_z+L_yL_z}$, an exponential-in-area subextensive degeneracy. The same model reduces to the 3D toric code when local 1-holonomy plaquette terms are added, so the fracton-like behavior is protected only by the subsystem symmetry, not topologically. For both the 2D and 3D models, the entanglement entropy of a subregion $A$ equals the logarithm of the ground-state degeneracy of a suitably restricted model, matching a formula previously established for topological models. A reader should care because this shows fracton-like order can appear in ordinary lattice gauge theories, not only in specially engineered stabilizer codes.","feed_headline":"Line symmetries in a 3D gauge model create fracton-like order","feed_subtitle":"Ground-state degeneracy grows like 2 to the boundary area; adding toric-code terms collapses it.","key_machinery":"The load-bearing objects are the cube operators $B_c^{(\\mu)}$ for $\\mu=x,y,z$, each defined as a projector that compares the holonomies of the two opposite plaquettes of cube $c$ orthogonal to $\\mu$; a ground state must satisfy all three at once, which forces any dual membrane representing a $\\mathbb{Z}_2$ flux to be straight and non-bending in the interior and to end on the boundary. Together with the vertex gauge projectors $A_v$, these operators form a commuting set, making the Hamiltonian exactly solvable. The degeneracy count then reduces to a boundary combinatorics problem: drawing independent dot configurations on each of the three pairs of opposite boundary faces, yielding $2^{L_xL_y+L_xL_z+L_yL_z}$.","core_discovery":"The central claim is that the Hamiltonian $H=-\\sum_v A_v-\\sum_c(B_c^{(x)}+B_c^{(y)}+B_c^{(z)})$ on a cubic lattice is exactly solvable by simultaneous diagonalization of its commuting projectors, and that its ground-state sector on a 3-ball is spanned by states represented by straight, non-bending dual membranes that start and end at opposite boundary faces. Counting independent boundary endpoints gives Eq. (17), $\\mathrm{GSD}=2^{L_xL_y+L_xL_z+L_yL_z}$. The elementary excitations are $\\mu$-fluxes located at membrane corners; they cannot move individually because moving one would require bending a membrane, which excites cube operators, whereas bound pairs can move along straight lines. The vertex charges are the same as in the 3D toric code and are fully mobile. Adding plaquette 1-holonomy terms to the Hamiltonian breaks the subsystem symmetry; in the limit of adding them everywhere, the ground-state sector becomes exactly that of the 3D toric code, so the model is a subsystem-symmetry-enriched topological phase rather than an intrinsically protected fracton phase.","pith_inferences":["Beyond the paper: on manifolds with nontrivial topology, the same dot-counting should combine with non-contractible closed membrane sectors, and one could check whether the total degeneracy is a sum or a product of the exponential term and the topological terms; the authors only identify the topological contribution qualitatively on the 3-torus.","Beyond the paper: the 2D model's connection to a classical eight-vertex-type model suggests a classical statistical-mechanical dual for the 3D model whose partition function would reproduce the area-law degeneracy, and computing that partition function on finite lattices would provide an independent check of Eq. (17).","Beyond the paper: the paper shows that symmetry-breaking local terms collapse the exponential degeneracy, but it does not analyze generic local perturbations that preserve the subsystem symmetry; testing stability under such symmetric perturbations would determine whether the exponential degeneracy survives as a subsystem-symmetry-protected feature."],"forward_implications":["On a 3-ball, the model has ground-state degeneracy $2^{L_xL_y+L_xL_z+L_yL_z}$, so the degeneracy grows exponentially with boundary area rather than with volume.","The spectrum contains fully mobile charge excitations identical to those of the 3D toric code, alongside $\\mu$-flux fractons at membrane corners that are immobile unless moved in pairs.","Adding 3D toric-code plaquette operators for all plaquettes reduces the ground-state degeneracy to the toric code's topological value, realizing a subsystem-symmetry-enriched topological phase.","For a subregion $A$, the entanglement entropy obeys $S_A=\\log(\\mathrm{GSD}_{\\tilde A})$, with the constant term equal to the 3D toric code's topological entanglement entropy.","The construction extends to arbitrary finite gauge groups, with commuting projectors whose algebra is the quantum double $D(G)$, offering a route to non-Abelian flux quasi-particles without dyons in the 2D case."],"supporting_citations":[{"why":"This is the baseline model whose vertex and plaquette operators are the building blocks, and whose ground states are contained in the new model's ground states.","marker":"[5]"},{"why":"This supplies the higher-gauge holonomy language (0- and 1-holonomy operators) used to define the cube operators as comparisons of opposite plaquette holonomies.","marker":"[22]"},{"why":"This frames fracton topological order as generalized lattice gauge theory, the standard construction that the paper's approach contrasts with.","marker":"[24]"},{"why":"This introduces the 2D hyperbolic fracton model whose subsystem-symmetry treatment is reviewed as the guide for the 3D model.","marker":"[51]"},{"why":"This companion treatment of the same 2D model via the dual eight-vertex model is used as a guide to the fracton-like properties of the 2D case.","marker":"[52]"},{"why":"This shows how gauging subsystem symmetries produces fracton phases and supplies the conceptual context for the subsystem symmetries used here.","marker":"[55]"},{"why":"This states the general relation between entanglement entropy and restricted ground-state degeneracy that the paper verifies for its models.","marker":"[62]"},{"why":"This introduces the original 2D plaquette model that Section II reviews as the two-dimensional fracton-like example.","marker":"[63]"},{"why":"This further develops that 2D model and its dimensional-reduction properties, grounding the review of the 2D case.","marker":"[64]"}],"fun_headline_variants":["Subsystem symmetries build fracton-like order in 3D gauge models","Fracton-like phase from line symmetries: mobile and immobile charges","3D gauge theory with subsystem symmetries shows fracton-like behavior","Subsystem symmetries yield fracton-like order with mobile charges"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The count in Eq. (17) assumes every ground state on the 3-ball is uniquely represented by an independent choice of straight membrane endpoints on each pair of opposite boundary faces, with no additional equivalence relations among those choices.","fun_headline_variants_meta":{"raw":{"variants":["Subsystem symmetries build fracton-like order in 3D gauge models","Fracton-like phase from line symmetries: mobile and immobile charges","3D gauge theory with subsystem symmetries shows fracton-like behavior","Subsystem symmetries yield fracton-like order with mobile charges"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000303,"raw_usage":{"total_tokens":1740,"prompt_tokens":941,"completion_tokens":799,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":557,"completion_tokens_details":{"reasoning_tokens":721}},"tokens_in":557,"tokens_out":799,"duration_ms":7628,"temperature":1.0,"reasoning_tokens":721,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T12:02:09.696064+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a small 3-ball lattice, say $L_x=L_y=L_z=2$, and directly enumerate all states satisfying every vertex and cube projector in the Hamiltonian. If the ground-state dimension is not $2^{12}=4096$ as Eq. (17) predicts, or if two different boundary dot configurations produce the same logical state, then the straight-membrane counting assumption fails.","supporting_citations":[{"cited_title":"Topological Order from a Cohomological and Higher Gauge Theory perspective","cited_arxiv_id":"1711.04186","evidence_quote":"This supplies the higher-gauge holonomy language (0- and 1-holonomy operators) used to define the cube operators as comparisons of opposite plaquette holonomies."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"This frames fracton topological order as generalized lattice gauge theory, the standard construction that the paper's approach contrasts with."},{"cited_title":"Weinstein, E","cited_arxiv_id":null,"evidence_quote":"This introduces the 2D hyperbolic fracton model whose subsystem-symmetry treatment is reviewed as the guide for the 3D model."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"This companion treatment of the same 2D model via the dual eight-vertex model is used as a guide to the fracton-like properties of the 2D case."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"This shows how gauging subsystem symmetries produces fracton phases and supplies the conceptual context for the subsystem symmetries used here."},{"cited_title":"Raussendorf, C","cited_arxiv_id":null,"evidence_quote":"This states the general relation between entanglement entropy and restricted ground-state degeneracy that the paper verifies for its models."},{"cited_title":"Devakul and D","cited_arxiv_id":null,"evidence_quote":"This introduces the original 2D plaquette model that Section II reviews as the two-dimensional fracton-like example."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"This further develops that 2D model and its dimensional-reduction properties, grounding the review of the 2D case."}],"review_version":1}