{"id":"669e83d8-e9db-4fdf-8892-3f672bb4bb52","arxiv_id":"2412.11097","paper_version":4,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The paper establishes a monitored analogue of bulk-edge correspondence: gapped area-law phases of Majorana circuits are separated by Lyapunov edge zero modes and a parity-based topological invariant, while the critical phase is gapless and dynamically characterized at O(L) timescales.","lead":"This paper shows that repeatedly measured chains of Majorana fermions have a topological character: in one phase a protected edge mode appears, and a parity-based invariant distinguishes that phase from a trivial one. It matters because measurement-induced phase transitions are a frontier in quantum many-body physics, and this work offers a way to define topology in monitored, nonequilibrium dynamics.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The invariant χ_T relies on convergence of the QR-constructed determinant det[eO_T] to the Pfaffian parity Q_T; this convergence is asserted but not demonstrated, and if it fails the bulk-edge discriminator could be a numerical convention artifact.","rationale":"The reader's weakest assumption—that Oseledec's theorem holds under the Born rule—is the right broad concern, and the paper itself acknowledges it as mathematically nontrivial. My stress test sharpens that concern to a specific unproved step that is more directly load-bearing: even if the Lyapunov spectrum exists, the topological invariant is defined through det[eO_T(s)], the determinant of the QR-constructed Lyapunov-vector matrix, and the paper only asserts its convergence to the Pfaffian parity Q_T(s). The supplement's numerical support shows convergence of the Lyapunov exponents for one trajectory and agreement with a sample average, but it does not compare det[eO_T] with the explicit Pfaffian formula, nor does it test sensitivity to the QR initial condition/convention. Without that check, the sharp separation of the two area-law phases by χ_T could be an artifact of the orientation of the Lyapunov basis rather than a genuine topological invariant. This is not an allegation of error; the numerics may well be correct. But the claimed bulk-edge correspondence stands or falls on this convergence, so a direct small-system comparison to Q_T is the decisive test. Since the reader already marked the paper CONDITIONAL and this concern is a concrete instance of the same convergence gap, I do not propose moving the verdict, hence UNCHANGED.","tokens_in":46002,"tokens_out":6204,"duration_ms":65928,"concrete_test":"For small system sizes (e.g., L=4, 6), pick Born-rule trajectories s at representative parameters (J=0 and 0.5; μ_o=0.1, 0.5, 0.9). For each trajectory and boundary condition, directly form K_T(s), compute H_eff,T(s)=-(i/2T)ln[K_T(s)K_T(s)^†] for moderate T (e.g., 10^2, 10^3, 10^4), and evaluate Q_T(s)=sgn(Pf[H_PBC_eff,T(s)]Pf[H_APBC_eff,T(s)]) by explicit Pfaffian. Compare with χ_T(s)=det[eO_PBC_T(s)]det[eO_APBC_T(s)] from the QR algorithm on the same sequence. Also re-run the QR construction with several independent initial W_0 and with different QR phase conventions. If χ_T and Q_T disagree in sign for a nonvanishing fraction of trajectories, or if det[eO_T] depends on the initial W_0 / convention, then the invariant is not well-defined and the central claim needs revision.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central bulk-edge claim rests on the invariant χ_T(s)=P_PBC_T(s)P_APBC_T(s) with P_T(s)=det[eO_T(s)], stated in Eqs. (7)-(8) to approximate and converge to Q_T(s)=sgn(Pf[H_PBC_eff,T(s)]Pf[H_APBC_eff,T(s)]). This convergence is asserted immediately after Eq. (7), but no proof is provided. For finite T, eO_T(s) is the matrix of QR-based Lyapunov vectors and is not orthogonal; its determinant depends on the QR convention and on the initial matrix W_0 used in the QR iteration. Oseledec's theorem, invoked after Eq. (4) and in Supplemental Sec. I.B, is not automatically applicable when the outcome sequence s is generated by the Born rule, since the matrix sequence is then state-dependent; the authors flag this as nontrivial and cite Refs. [66,67], but the only supporting evidence is Supplemental Fig. S1, which shows a single-trajectory measurement-only-style example at L=4, J=0.5, μ_o=0.5 and compares it to a 100-trajectory average. That does not establish that det[eO_T(s)] concentrates on the same sign as Pf[H_eff,T(s)] for both PBC and APBC. If det[eO_T] does not converge to the parity sign, then the claimed distinction between the topological (χ=-1) and trivial (χ=+1) area-law phases, and hence the bulk-edge correspondence itself, is not quantitatively established.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This Letter proposes a Lyapunov-analysis framework for monitored Majorana circuits and uses it to claim a bulk-edge correspondence in the gapped measurement-induced phases: the topological area-law phase is distinguished by edge-localized Majorana zero modes within the bulk Lyapunov gap, the trivial area-law phase has no such modes, and a fermion-parity invariant χ_T(s) = P_T^PBC(s) P_T^APBC(s), built from QR-based Lyapunov vectors, takes χ = -1 (+1) in the topological (trivial) phase. The antiperiodic boundary condition is defined by twisting only the boundary measurement outcomes relative to a PBC trajectory, with the unitary boundary term J' = -J. The critical sub-volume-law phase is claimed to be gapless in the Lyapunov spectrum and is dynamically characterized by χ evaluated at T = O(L). The paper is transparent about its main assumptions: applicability of Oseledec's theorem to Born-rule-generated outcome sequences is flagged as mathematically nontrivial (main text after Eq. (4); Supp. Sec. I.B), and the critical-phase scaling is stated to be inconclusive at available sizes (Supp. Sec. III).","tokens_in":46374,"tokens_out":12633,"duration_ms":112824,"significance":"If the central claim holds, the paper is a substantive step: it extends bulk-edge correspondence, a cornerstone of equilibrium topology, to monitored dynamics and supplies a concrete invariant (the PBC/APBC parity product with twisted boundary measurement outcomes) that sharply separates the two area-law phases. The paper has notable strengths: the APBC construction is a new idea; the Lyapunov-gap criterion gives falsifiable predictions (edge-mode localization, χ = ±1, gapless critical spectrum); code and data are shared (Ref. [91]); and no free parameters are fitted to the target result, since the model parameters are scanned. The authors are also unusually explicit about which steps are numerically checked rather than proved. The main weakness is that the convergence of the QR-based determinant to the Pfaffian parity, on which the invariant rests, is asserted rather than demonstrated.","major_comments":[{"comment":"The statement that χ_T(s) 'approximates and converges to Q_T(s) for T → ∞' is the linchpin of the phase discrimination, but it is not demonstrated. For finite T, Õ_T(s) is not orthogonal, and det[Õ_T(s)] depends on the QR convention and on the random initial matrix W_0 used in the QR iteration; convergence of the Lyapunov exponents (Supp. Fig. S1) does not by itself establish that sign(det[Õ_T]) stabilizes to the Kitaev parity for both PBC and APBC. The trajectory-averaged χ_T in Figs. 2(e) and 3(d) shows that the sign is stable across outcomes, but it does not test equality with Q_T(s) = sgn(Pf[H_eff,T^PBC] Pf[H_eff,T^APBC]) or invariance under re-initialization of W_0. I ask for (i) a direct comparison of sign(det[Õ_T(s)]) with Q_T(s) for small L and T, where Pf[H_eff,T(s)] is computable; (ii) a check that the sign is unchanged under different random W_0 and QR conventions; and (iii) both checks in each gapped phase and boundary condition.","section":"Topological invariant and bulk-edge correspondence, Eqs. (7)-(8)"},{"comment":"The entire Lyapunov construction rests on Oseledec's theorem for products K_T(s) whose factors depend on the Born-rule-generated outcome sequence s. The authors correctly flag the Born-rule complication as mathematically nontrivial and cite Refs. [66,67], but the supporting numerics in Supp. Fig. S1 are a single trajectory at one parameter set (L=4, J=0.5, μ_o=0.5) compared with a 100-trajectory average. Since the edge-mode and invariant claims are conditional on this convergence, and since convergence could in principle fail in some phases, I request additional single-trajectory-versus-ensemble convergence checks in all three phases and under OBC, PBC, and APBC, or a more prominent statement that the bulk-edge conclusions are conditional on this numerically checked assumption.","section":"Model and Lyapunov analysis, after Eq. (4); Supp. Sec. I.B"},{"comment":"The abstract and summary state that the critical phase has 'a bulk gapless spectrum in the critical phase,' but the evidence is the decay of z1 with L in the window 0.4 ≲ μ_o ≲ 0.6 (Fig. 3 and Fig. S3), with scaling close to 1/L, which the authors themselves describe as inconclusive ('we cannot conclude the true scaling form especially inside the critical phase'). A 1/L decay of the lowest Lyapunov exponent is consistent with gaplessness, but the claim as stated is stronger than the data. Either the claim should be softened to 'consistent with a gapless phase' or a more conclusive scaling analysis should be provided, for example with subleading corrections or an independent probe such as the (ln L)^2 entanglement scaling already used to identify the phase.","section":"Critical gapless phase with additional unitary; Supp. Sec. III"},{"comment":"The APBC is defined by flipping only the boundary outcomes s_{2L,t} relative to a PBC trajectory, so the APBC circuit is not sampled from the APBC Born rule (the postselection overhead is acknowledged). This is a legitimate construction, but the paper gives no argument, beyond the numerical sharpness of Fig. 2(e), that this particular twist is the correct monitored analogue of antiperiodic boundary conditions. Because χ_∞ = -1/+1 in the two gapped phases is the central quantitative claim, I ask for a robustness check of the invariant under alternative twists (for example, flipping a different boundary degree of freedom, or using only the unitary twist J' = -J without flipping outcomes) and a brief justification of why the flipped-outcome twist tracks the parity sector in which the APBC gap closes at the transition.","section":"Topological invariant and bulk-edge correspondence, APBC definition"}],"minor_comments":[{"comment":"The stopping criterion 'less than √10 × 10^{-3}' is unexplained; the factor √10 appears to be a typo, and the criterion should be stated plainly.","section":"Supp. Sec. I.B, stationarity criterion"},{"comment":"The sentence 'P_T^PBC(s) = 1 is satisfied in the whole parameter region' reads like an exact identity, although the following sentence explains it numerically via the finite-size splitting z1 ~ 1/L; please rephrase as a numerical observation for the finite systems studied.","section":"Paragraph after Eq. (8)"},{"comment":"The caption states 'after a sufficiently long time,' but the temporal-averaging and stopping criteria are given only in the supplement; please state the criterion in the caption or in the main text.","section":"Fig. 2(e) caption"},{"comment":"Minor grammar: 'we will also call χT (s) as the topological invariant' should read 'we will also call χT(s) the topological invariant.'","section":"Sentence following Eq. (8)"},{"comment":"The abstract promises 'a general framework' for monitored quantum systems, while the demonstration is for free-fermion Gaussian states; the outlook acknowledges this, but the abstract should qualify the claim, for example as a framework demonstrated on free-fermion monitored systems with a proposed route to interacting systems.","section":"Abstract"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is a credible Letter candidate, but the convergence of χ_T to the Pfaffian parity is a load-bearing point that is currently asserted rather than checked, and the Oseledec/Born-rule numerical support is thin (one parameter point). Both issues are fixable with targeted numerics, so major revision rather than rejection seems appropriate. The near-simultaneous related work (Ref. [90]) is disclosed and coordinated; the present paper's distinct contribution is the Lyapunov spectrum/edge-mode analysis and the APBC parity invariant. The self-citations (Refs. [57,60]) are method references and do not raise circularity concerns."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"You should know this paper is a genuine step forward: it gives a concrete Lyapunov-based framework for monitored Majorana circuits, shows edge-localized zero modes in the topological area-law phase, and constructs a parity invariant that cleanly separates the two gapped phases. The twisted-APBC definition is a real conceptual contribution, and the numerics are coherent, with code and data shared on Zenodo. The authors also honestly flag where their evidence is not conclusive, which is more than many papers do.\n\nThe main result — that gapped monitored phases obey a bulk-edge correspondence — holds up well as a numerical claim. The zero-mode localization and the χ = −1 versus +1 separation are demonstrated across system sizes up to L=128. The critical-phase story is weaker but the authors are careful to say that the scaling form cannot be definitively determined; that is a limitation, not a flaw.\n\nThe softest spot is the convergence of χ_T (the determinant of the QR-based Lyapunov-vector matrix) to the Pfaffian parity Q_T. This is asserted rather than proved, and for finite T the determinant does depend on the QR convention and initial vectors. I agree with the stress-test that this is a gap. But I would not call it a likely fatal flaw: in the limit, the Lyapunov-vector matrix diagonalizes the effective Hamiltonian, and the determinant of that matrix is exactly the sign of the Pfaffian up to positive factors. The numerical time series in Fig. S4 show χ_T converging to ±1, so the concern is about rigor, not about a demonstrated artifact.\n\nThe other issue is novelty. The note added acknowledges overlap with Ref. [90], and Ref. [42] already discusses topological modes in monitored dynamics. The paper does not spell out precisely which of its results go beyond those two works. That needs to be clarified in a revision.\n\nWho should read this: anyone working on measurement-induced phases, topological dynamics in free-fermion circuits, or Lyapunov methods for nonunitary evolution. It deserves a serious referee. With a proof or a much stronger argument for the convergence of χ_T, and an explicit comparison to Refs. [42,90], I would be happy to see it published.","headline":"A numerically convincing extension of bulk-edge correspondence to monitored Majorana circuits via Lyapunov analysis and a twisted-boundary parity invariant, with two rigor gaps that a revision should close.","tokens_in":46857,"tokens_out":2329,"would_cite":true,"duration_ms":25009,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Monitored Majorana circuits obey a bulk-edge correspondence in gapped phases.","keywords":["measurement-induced phase transitions","Majorana zero modes","Lyapunov spectrum","bulk-edge correspondence","monitored quantum circuits","fermion parity invariant","entanglement phases"],"falsifier":"Look at the lowest Lyapunov exponent under open boundary conditions at parameters inside the claimed topological area-law phase: if it remains finite as the system size grows, or if the corresponding mode's squared weight at the edges does not approach 1, then the edge zero mode does not exist and the bulk-edge correspondence claim collapses.","tokens_in":45814,"feed_emoji":"⚛️","tokens_out":6181,"duration_ms":51829,"temperature":0.7,"pith_summary":"This paper argues that topological order can be defined for monitored quantum dynamics, not only for ground states of static Hamiltonians. In a (1+1)-dimensional monitored Majorana circuit, it identifies the topological area-law phase by the presence of edge-localized Majorana zero modes inside an open bulk Lyapunov gap, and the trivial area-law phase by their absence. It constructs a fermion-parity invariant $\\chi_T(s)$ that takes different values in the two gapped phases and dynamically marks the critical phase. The critical phase is shown to be gapless in the Lyapunov spectrum. If correct, this extends the bulk-edge correspondence to a class of driven, measurement-determined systems.","feed_headline":"Monitored Majorana circuits obey a bulk-edge correspondence","feed_subtitle":"Lyapunov zero modes and a fermion-parity invariant separate topological, trivial, and critical phases.","key_machinery":"Lyapunov analysis of the random matrix product $K_T(s)$ governing the evolution of Majorana operators: the spectrum of the effective Hamiltonian $H_{\\mathrm{eff},T}(s) = -\\frac{i}{2T}\\ln[K_T(s)K_T^\\dagger(s)]$ converges to the Lyapunov spectrum, whose lowest non-negative exponent serves as the bulk gap. The corresponding Lyapunov vectors form an orthogonal matrix $O_T(s)$ that diagonalizes the effective Hamiltonian. The topological invariant is built from determinant products $\\det[O_T^{\\mathrm{PBC}}(s)]$ and $\\det[O_T^{\\mathrm{APBC}}(s)]$, with the antiperiodic boundary condition defined by twisting only the boundary measurement outcomes relative to the periodic trajectory; this is a fermion-parity difference generalizing the static Majorana-chain construction.","core_discovery":"The central claim is that monitored Majorana circuits in gapped phases satisfy a bulk-edge correspondence. The paper defines an effective Hamiltonian from the long-time Lyapunov analysis of the random matrix product that evolves Majorana operators, and shows that in the topological area-law phase the lowest Lyapunov exponent under open boundaries vanishes and the corresponding mode is localized at the edges, while in the trivial area-law phase it stays finite and delocalized. A bulk topological invariant is obtained by multiplying the fermion parities of the ground states of the effective Hamiltonian under periodic and antiperiodic boundary conditions, where the antiperiodic circuit is defined by flipping the boundary measurement outcomes of the periodic trajectory; this invariant is $\\chi_T(s) = P_{\\mathrm{PBC}}(s) P_{\\mathrm{APBC}}(s)$, converging to $\\pm 1$ in the two gapped phases. In the critical sub-volume-law phase the bulk Lyapunov gap closes, and the same invariant characterizes the phase dynamically through its slow relaxation at times $T = O(L)$.","pith_inferences":["The same Lyapunov edge-mode diagnostic should work for other monitored free-fermion circuits, but the twisted-boundary outcome prescription may need modification when measurements have more than two outcomes.","Because evaluating the antiperiodic trajectory requires postselecting on the periodic outcome sequence, direct large-scale measurement of $\\chi_T$ is exponentially costly; a cheaper experimental proxy might be the slow relaxation of parity fluctuations near the critical phase.","If the assumption that Oseledec's theorem applies to Born-rule trajectories is eventually proven, the Lyapunov spectrum could serve as a rigorous classification tool for measurement-induced topological phases rather than a numerical diagnostic."],"forward_implications":["In the topological area-law phase, open-boundary monitored circuits host edge-localized Majorana zero modes inside the bulk Lyapunov gap.","In the trivial area-law phase, no such edge modes appear, so the two gapped phases are distinguished by both the Lyapunov spectrum and the parity invariant.","The critical phase has a gapless Lyapunov spectrum, and the invariant fails to converge on the $O(L)$ timescale, giving a dynamical signature of the two transition points.","The framework extends the bulk-edge correspondence to monitored dynamics and suggests how to define topological invariants for symmetry-protected monitored phases."],"supporting_citations":[{"why":"Provides the static Majorana-chain construction whose zero modes and parity invariant the monitored setting adapts.","marker":"[68]"},{"why":"Supplies the multiplicative ergodic theorem and QR-based numerical method used to define the Lyapunov spectrum and vectors.","marker":"[63-65]"},{"why":"Gives the invariant-measure results that the paper relies on for applying Oseledec's theorem to Born-rule monitored trajectories.","marker":"[66,67]"},{"why":"Establishes the continuous-time free-fermion monitored model that predicts the three entanglement phases whose locations this paper compares with.","marker":"[53]"},{"why":"Provides prior numerical evidence for topological transitions in weakly monitored free fermions, the baseline for the area-law phase structure.","marker":"[41]"},{"why":"Identifies topological modes in monitored quantum dynamics, supporting the presence of distinct area-law phases.","marker":"[42]"}],"fun_headline_variants":["Monitored Majorana circuits show bulk-edge correspondence","Lyapunov modes define topology in monitored Majorana systems","Bulk-edge invariant distinguishes monitored Majorana phases","Measurement-induced phases: topology from Lyapunov spectrum","Majorana monitored: bulk-edge from Lyapunov analysis"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument assumes that the random matrix product built from a monitored circuit has a well-defined Lyapunov spectrum that is independent of the particular measurement-outcome sequence; the paper provides numerical support but no proof in the Born-rule setting.","fun_headline_variants_meta":{"raw":{"variants":["Monitored Majorana circuits show bulk-edge correspondence","Lyapunov modes define topology in monitored Majorana systems","Bulk-edge invariant distinguishes monitored Majorana phases","Measurement-induced phases: topology from Lyapunov spectrum","Majorana monitored: bulk-edge from Lyapunov analysis"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000405,"raw_usage":{"total_tokens":2094,"prompt_tokens":917,"completion_tokens":1177,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":533,"completion_tokens_details":{"reasoning_tokens":1111}},"tokens_in":533,"tokens_out":1177,"duration_ms":9559,"temperature":1.0,"reasoning_tokens":1111,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T15:18:19.623160+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Look at the lowest Lyapunov exponent under open boundary conditions at parameters inside the claimed topological area-law phase: if it remains finite as the system size grows, or if the corresponding mode's squared weight at the edges does not approach 1, then the edge zero mode does not exist and the bulk-edge correspondence claim collapses.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes the continuous-time free-fermion monitored model that predicts the three entanglement phases whose locations this paper compares with."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Identifies topological modes in monitored quantum dynamics, supporting the presence of distinct area-law phases."}],"review_version":1}