{"id":"c0c4d93c-c241-44c7-a016-28d5a5aeff4f","arxiv_id":"2501.02398","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"For 2D spin systems with mixed-state approximate Haag duality, finite-depth decoherence maps the anyon category of the final state faithfully into that of the initial state, so topological order cannot grow under local noise.","lead":"An operator-algebraic framework assigns a braided tensor category of anyons to two-dimensional quantum spin systems in mixed states, provided the states satisfy a technical locality condition called approximate Haag duality. The paper proves that applying a finite-depth quantum channel cannot create new anyon species: the final state's category embeds faithfully into the initial state's category.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 1.12's decisive step is imported from [Oga22] without adapting it to mixed-state partial isometries, and Definition 1.1 is never instantiated, so the central implication is unverified in the regime it claims.","rationale":"The reader flagging Definition 1.1 is correct that the theorem has no demonstrated physical instances. However, the more immediately load-bearing issue for the central claim is that the proof of Theorem 1.12 delegates the decisive category equivalence under approximately factorizable automorphisms to the pure-state paper [Oga22]. The mixed-state objects are structurally different from the pure-state ones: they are implemented by partial isometries whose initial and final supports are a possibly nontrivial central projection rho(I), and the morphism spaces are cornered by rho(I), sigma(I). The pure-state proofs typically use unitarity of intertwiners in essential ways, and the paper gives no verification that those arguments survive when V*V=rho(I) is not the identity. This is not an accusation of fraud; it is a concrete missing proof in a central step. The reader's weaker-assumption point about Definition 1.1 remains important: even if the missing equivalence is supplied, the classification statement applies only to states satisfying that definition, and none are constructed. Together these considerations justify keeping the paper conditional rather than accepting the main theorem as fully supported at this stage. The reader's verdict of CONDITIONAL is therefore the right one; I would not change it, though I would phrase the primary concern as the unproved adaptation of the automorphism-invariance proof rather than only the lack of examples.","tokens_in":28542,"tokens_out":11452,"duration_ms":123962,"concrete_test":"Write out the missing Lemma: for \\tilde C_{omega,Λ0} (Definition 1.6), with alpha approximately factorizable, construct an explicit braided equivalence \\tilde C_{omega,Λ0} ≅ \\tilde C_{omega∘alpha,Λ0}, keeping V*V=rho(I) rather than I at every step. If the construction from [Oga22] requires V to be unitary in order to define the inverse sector or to prove that the braiding intertwines, then the proof of Theorem 1.12 does not transfer to the mixed-state setting and the theorem as stated is unproved.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The main theorem is the conditional implication omega2=(omega1⊗psi1)∘alpha|A ⇒ C_{omega2⊗psi2,Λ0} ↪ C_{omega1⊗psi2,Λ0}. Its proof in Section 5 rests on the sentence 'By the same proof as [Oga22]' for the equivalence of \\tilde C_{omega1⊗psi2⊗psi1,Λ0} and \\tilde C_{(omega1⊗psi2⊗psi1)alpha13,Λ0}. That equivalence is the step that converts an approximately factorizable automorphism into a category isomorphism, and it is not a special case of any theorem stated or proved here. [Oga22] treats pure states whose sectors are implemented by unitaries V with V*V=I; here the objects of \\tilde C (Definition 1.6) are implemented by partial isometries with V*V=VV*=rho(I) a possibly nontrivial central projection, and morphisms are constrained by sigma(I)Rho(I)=R. Unitarity is used in the standard proof of such invariance, for example in defining the inverse functor and in checking that intertwiners on shifted cones compose to a unitary on the original cone. No argument is given that every step survives with rho(I) not equal to 1. Independently, Definition 1.1 is the only hypothesis that makes the category nonempty, yet the paper gives no concrete non-pure state satisfying it; Lemma 1.2 only stabilizes proper infiniteness, Lemma 4.4 and Lemma 5.2 only propagate the assumption to tensor products and automorphic images, and finite-temperature Gibbs states or decohered toric-code states are not checked. If Definition 1.1 has no such instances, Theorem 1.12 is vacuous for the advertised channel-decoherence scenario.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes an operator-algebraic framework for assigning braided C*-tensor categories to mixed states of two-dimensional quantum spin systems. The key new object is the category C_{omega, Lambda0} of superselection sectors for a state omega satisfying a mixed-state version of approximate Haag duality (Definition 1.1), with objects implemented by partial isometries after stabilization by a pure infinite tensor product state. The paper proves a stabilization theorem (Theorem 1.9) saying that the category is unchanged, up to equivalence, when additional pure tensor factors are added, and it constructs a faithful braided tensor functor from the category of a subsystem to the category of the full system (Theorem 3.2). The main result (Theorem 1.12) asserts that if omega2 is obtained from omega1 by an approximately factorizable automorphism on the composite system, then C_{omega2 tensor psi2, Lambda0} embeds faithfully as a braided tensor subcategory of C_{omega1 tensor psi2, Lambda0}; this is interpreted as a monotonicity statement for anyonic data under decoherence by finite-depth quantum channels.","tokens_in":28931,"tokens_out":5277,"duration_ms":55115,"significance":"If the categorical construction and the invariance under approximately factorizable automorphisms can be fully justified in the partial-isometry setting, the paper would provide a rigorous language for mixed-state topological order and a general monotonicity principle for anyonic sectors under decoherence. The author proves the stabilization lemmas in considerable detail, including the technically nontrivial Lemma 4.3, and the subsystem functor of Section 3 is explicit and checkable. The paper contains no fitted parameters and no hidden axioms beyond the stated approximate Haag duality. However, the central braided tensor category is not fully constructed in the text for the mixed-state case, and the decisive invariance step in the proof of Theorem 1.12 is imported from [Oga22] without an adaptation argument. In addition, no concrete mixed state is shown to satisfy Definition 1.1, so the physical applicability of the main theorem is not yet demonstrated.","major_comments":[{"comment":"The sentence 'By the same proof as [Oga22], the braided C*-tensor categories ... are equivalent' is the decisive step that turns an approximately factorizable automorphism into a category equivalence, and it is not a special case of any theorem proved in this paper. In [Oga22] the superselection sectors are implemented by unitaries with V*V = VV* = I, whereas in Definition 1.6 the objects of \\tilde C are implemented by partial isometries with V*V = VV* = rho(I), where rho(I) can be a nontrivial central projection. The standard proof of invariance under approximately factorizable automorphisms uses unitarity in essential places, for example in constructing the inverse functor and in checking that intertwiners on shifted cones combine to a unitary on the original cone. The author should either prove that every step of the [Oga22] argument survives with rho(I) not equal to 1 or formulate and prove a separate lemma for the mixed-state categories.","section":"Section 5, proof of Theorem 1.12"},{"comment":"The construction of the braided C*-tensor category C_{omega, Lambda0} is delegated to [Oga22] and [Oga24] with the statement that the proof is the same except for the existence of subobjects, for which Lemma 2.2 is supplied. This leaves the tensor product, braiding, direct sums, and the verification that the morphism spaces define a C*-category essentially unverified in the mixed-state setting. Since the tensor product formula (2.4) and braiding formula (2.5) are load-bearing for the main theorem, the paper should contain a precise statement of how the partial-isometry objects satisfy each structural requirement, or explicitly identify the exact theorem in the prior literature that covers this generalization.","section":"Section 2, Theorems 1.8 and 2.1"},{"comment":"No concrete non-pure state is exhibited that satisfies the mixed-state approximate Haag duality of Definition 1.1. Lemma 1.2 only stabilizes proper infiniteness, while Lemmas 4.4 and 5.2 propagate the assumption to tensor products and automorphic images. Without a single example -- for instance a finite-temperature Gibbs state, a decohered toric-code state, or a state obtained by a concrete channel from a known pure state -- the hypotheses of Theorem 1.12 may define an empty class, and the theorem would not apply to the channel-decoherence setting advertised in the abstract. The author should add an example or explicitly state as an open problem whether physically relevant mixed states satisfy Definition 1.1.","section":"Definition 1.1 and Section 5"}],"minor_comments":[{"comment":"The inequality 'vv*/lessnotequal I' is a typographical artifact and should read 'vv* != I'; similarly, the notation 'z⊗v' and 'z⊗ I' should be made precise, since z is a central projection in the first tensor factor and v is an isometry in the second.","section":"Lemma 1.2"},{"comment":"The condition 'V_{\\sigma\\Lambda_2(t)} \\in V_{\\sigma\\Lambda_2(t)}' is self-referential as printed; it should be 'V_{\\sigma,\\Lambda_2(t)} \\in \\mathcal{V}_{\\sigma,\\Lambda_2(t)}' or another unambiguous notation for the set of partial isometries associated to the shifted cone.","section":"Theorem 2.1(iv)"},{"comment":"The sentence 'defines a equivalence H between the braided C*-tensor functors \\tilde C_{...} and \\tilde C_{...}' appears to say 'equivalence' between categories, not functors; the wording should be corrected.","section":"Corollary 1.10 proof"},{"comment":"The notation B^{(1)}_1, B^{(2)}_1, K^{(1)}_1, K^{(2)}_1 is technically correct but difficult to parse; a short accompanying sentence explaining the factorizations (4.38) and (4.39) and their role in isolating the B^{(1)}_1 tensor factor would improve readability.","section":"Lemma 4.3"}],"recommendation":"major_revision","confidential_remarks":"The main theorem is conditional on an unproved import from [Oga22] and on the non-emptiness of Definition 1.1. I recommend that the editor require the author to either prove the invariance under approximately factorizable automorphisms for partial-isometry sectors or supply a precise theorem statement with a full proof, and to address the absence of concrete examples. The paper is otherwise a coherent framework with substantial technical content, and the requested revisions appear feasible within the scope of the manuscript."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper is worth reading. It defines a braided C*-tensor category for mixed states satisfying a mixed-state approximate Haag duality, and it proves the monotonicity result: under an approximately factorizable automorphism (hence under finite-depth channels via Stinespring), the category of the final state embeds faithfully into the category of the initial state. The physical message is the right one: decoherence can split or lose anyons, but it cannot create them.\n\nWhat is genuinely good: the subsystem functor in Theorem 3.2 is proved in detail and is faithful; Lemma 3.3 gives full faithfulness when the extra tensor factor is pure. Lemma 2.2 on subobjects is a real addition, and the equivalence between the full category and the stabilized subcategory is handled explicitly. The paper is honest that the embedding is not full, so splitting is allowed. There is no circularity and no fitted parameters: the result is conditional, but the condition is stated exactly.\n\nThe soft spots are two. First, Definition 1.1, the mixed-state approximate Haag duality, is never instantiated for any concrete non-pure state. The paper proves the condition is preserved under stabilization and automorphic images, but no finite-temperature Gibbs state, decohered toric code, or other physically natural mixed state is shown to satisfy it. Until that happens, Theorem 1.12 is a well-built conditional statement about a class that might be empty. This is an open problem rather than a proof error, but it is the main limitation. Second, the proof of Theorem 1.12 delegates the key equivalence under approximately factorizable automorphisms to \"the same proof as [Oga22]\". The objects here involve partial isometries with V*V = VV* = rho(I), where rho(I) is a possibly nontrivial central projection, and unitarity is used in the standard proof of such invariance. The paper gives no argument that every step survives this change. It may be routine—the techniques in Section 4 suggest it is—but a referee should ask for the details.\n\nThe paper is for people working in algebraic quantum field theory and rigorous aspects of topological order in open systems. My verdict is CONDITIONAL, not reject. I would send it to a serious referee, with a request that the omitted equivalence be written out or reduced to an explicit lemma, and that the author provide at least one concrete mixed state satisfying Definition 1.1. Engage with it.","headline":"A well-built conditional theorem: mixed-state anyon categories can only shrink under finite-depth channels, but the new duality condition is never instantiated and one key proof is imported from prior work.","tokens_in":29417,"tokens_out":2911,"would_cite":true,"duration_ms":30265,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["46L60","46L10","81T05","82B20"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper defines braided C*-tensor categories for mixed states of 2D quantum spin systems and proves that finite-depth decoherence makes the final state's category a subcategory of the initial state's.","keywords":["mixed-state topological order","braided C*-tensor category","approximate Haag duality","decoherence","finite-depth quantum channel","superselection sectors","two-dimensional quantum spin systems","operator algebras"],"falsifier":"Take a concrete mixed state that the theory should describe, for instance a finite-temperature Gibbs state of a local Hamiltonian or the toric code after local depolarizing noise, and check the approximate Haag duality inequalities of Definition 1.1 directly. If such a state violates the inequalities, the category $C_{\\omega\\otimes\\psi,\\Lambda_0}$ is not defined for it and the main theorem imposes no constraint on that decoherence process.","tokens_in":28348,"feed_emoji":"🌀","tokens_out":10213,"duration_ms":92143,"temperature":0.7,"pith_summary":"The paper extends the operator-algebraic classification of topological order from pure gapped ground states to mixed states. It attaches to any state satisfying a mixed-state version of approximate Haag duality a braided $C^*$-tensor category whose objects are the superselection sectors (anyon species) localized in cones. The main theorem says that if a finite-depth quantum channel takes one such state to another, the category of the final state is a braided $C^*$-tensor subcategory of the category of the initial state, though not necessarily a full subcategory. This gives a rigorous sense in which decoherence can fuse or identify anyons but cannot create new ones.","feed_headline":"Decoherence can only shrink a mixed state's anyon category","feed_subtitle":"Finite-depth noise embeds the final state's anyon category into the initial one, so no new anyon types appear.","key_machinery":"The load-bearing machinery is the mixed-state approximate Haag duality of Definition 1.1, together with the superselection criterion of Definition 1.3. Approximate Haag duality says that, up to unitaries that can be approximated by unitaries supported far away, the commutant of the algebra of the complement of a cone is contained in the double commutant of a slightly thickened cone; this is what lets localized representations be organized into a braided $C^*$-tensor category. The central object is the category $C_{\\omega\\otimes\\psi,\\Lambda_0}$, built by the standard superselection-sector recipe: objects are representations $\\rho$ of the spin algebra that agree with the GNS representation $\\pi$ outside cones, morphisms are intertwiners in $\\pi(A)''$, the tensor product is composition of endomorphisms, and the braiding is the limit $\\varepsilon(\\rho,\\sigma)=\\lim_{t\\to\\infty} V_{\\sigma,\\Lambda_2(t)}\\, T_{\\rho}(V_{\\sigma,\\Lambda_2(t)}^*)$, which measures the statistics of moving one sector around another. The stabilization by the pure infinite tensor product state $\\psi$ makes all cone von Neumann algebras properly infinite, a condition needed for the superselection criterion to work.","core_discovery":"The central discovery is that a mixed state can carry an anyon theory: to each state $\\omega$ satisfying the mixed-state approximate Haag duality and having properly infinite cone algebras, one can associate a braided $C^*$-tensor category $C_{\\omega\\otimes\\psi,\\Lambda_0}$, after tensoring with a pure infinite tensor product state $\\psi$ to stabilize. This category is independent of the stabilizer and of the reference cone up to equivalence. The main theorem states that if $\\omega_2$ is obtained from $\\omega_1$ by composing with an approximately factorizable automorphism of a larger system—equivalently, by applying a finite-depth quantum channel to the subsystem—and both states satisfy the duality assumption, then there is a faithful braided tensor functor from $C_{\\omega_2\\otimes\\psi_2,\\Lambda_0}$ to $C_{\\omega_1\\otimes\\psi_2,\\Lambda_0}$. In physical terms, after decoherence two anyons that looked different in the final state may become isomorphic in the original state, so the final anyon theory is a sub-theory, not necessarily a full sub-theory, of the initial one.","pith_inferences":["If finite-temperature or decohered topological states do satisfy the mixed-state approximate Haag duality, the theorem would provide a monotonicity law: braided tensor structure can only be lost, not gained, along finite-depth channels.","A natural next step is to compute the category for a concrete decohered model, for example the toric code under local noise, and see whether the resulting subcategory is strictly smaller yet not full; that would test how much information the non-fullness carries.","Because the invariant is defined only after stabilizing with a pure infinite tensor product state, a direct invariant of the physical mixed state alone would require an additional argument or a modified superselection criterion.","The faithfulness of the functor suggests a possible partial order on mixed-state topological orders generated by finite-depth channels; if the categories of two states each embed into the other, they would be equivalent, giving a candidate classification principle."],"forward_implications":["Under the theorem's assumptions, any superselection sector of the final state is already a sector of the initial state: decoherence cannot create new anyon species.","The embedding is faithful but not necessarily full, so distinct anyon types in the final state can become isomorphic in the initial state; decoherence can identify anyons that were separate.","Choosing a different reference cone or a different stabilizing pure state changes the category only by an equivalence, so the invariant is intrinsic to the state.","Automorphisms generated by local interactions and finite-depth quantum circuits are approximately factorizable, which connects the category statement to physically realistic finite-time interactions with an environment.","For pure states satisfying the assumptions, the new category agrees with the previously established pure-state category, so the mixed-state construction extends rather than replaces the pure-state invariant."],"supporting_citations":[{"why":"Supplies the pure-state category construction and the equivalence under approximately factorizable automorphisms that the mixed-state proof imports.","marker":"[Oga22]"},{"why":"Introduces relative Haag duality, whose approximate version Definition 1.1 adapts to mixed states.","marker":"[Cam07]"},{"why":"Establishes the operator-algebraic description of anyons in quantum double models that this paper generalizes to mixed states.","marker":"[Naa11]"},{"why":"Provides the localized-representation superselection recipe used to define objects and morphisms.","marker":"[DHR71]"},{"why":"Supplies the braid-statistics construction that defines the braiding on the category.","marker":"[FG90]"},{"why":"Gives the braid group statistics and exchange algebra framework used in organizing sectors into a braided tensor category.","marker":"[FRS89]"},{"why":"Shows automorphisms generated by local interactions are approximately factorizable, linking the main theorem to finite-depth channels.","marker":"[Oga21]"},{"why":"Provides the von Neumann algebra comparison theorems for properly infinite projections used in stabilization and subobject arguments.","marker":"[KR86]"},{"why":"Gives the representation-extension result used to build representations of the tensor product system from commuting subalgebras.","marker":"[BO08]"}],"fun_headline_variants":["Decoherence shrinks anyon categories in mixed states","Mixed states take anyon order, and noise only shrinks it","Finite-depth noise cannot create new anyon types","Anyon theory of a mixed state survives decoherence as a subtheory","Mixed-state anyon categories shrink under finite-depth channels"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The construction and the main theorem apply only to states that satisfy the mixed-state approximate Haag duality of Definition 1.1, and the paper does not prove that any concrete physically relevant mixed state, such as a finite-temperature Gibbs state or a decohered toric code state, satisfies it; if that assumption is empty, the category invariant never applies to the decoherence processes it aims to classify.","fun_headline_variants_meta":{"raw":{"variants":["Decoherence shrinks anyon categories in mixed states","Mixed states take anyon order, and noise only shrinks it","Finite-depth noise cannot create new anyon types","Anyon theory of a mixed state survives decoherence as a subtheory","Mixed-state anyon categories shrink under finite-depth channels"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000177,"raw_usage":{"total_tokens":1253,"prompt_tokens":864,"completion_tokens":389,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":480,"completion_tokens_details":{"reasoning_tokens":304}},"tokens_in":480,"tokens_out":389,"duration_ms":3935,"temperature":1.0,"reasoning_tokens":304,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T22:14:52.509268+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a concrete mixed state that the theory should describe, for instance a finite-temperature Gibbs state of a local Hamiltonian or the toric code after local depolarizing noise, and check the approximate Haag duality inequalities of Definition 1.1 directly. If such a state violates the inequalities, the category $C_{\\omega\\otimes\\psi,\\Lambda_0}$ is not defined for it and the main theorem imposes no constraint on that decoherence process.","supporting_citations":[],"review_version":1}