{"id":"2fd64948-96f0-44a6-a5a2-4f2b7c5083a9","arxiv_id":"2604.00358","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The half-chain waiting-time distribution in a continuously monitored Heisenberg chain develops a non-Poissonian tail whose decay rate is given by the dominant eigenvalue of the no-jump superoperator L0.","lead":"This paper shows that the random waiting times between quantum jumps in monitored quantum matter look different when you only watch half the system: a heavy exponential tail appears instead of the usual simple pattern. The result offers a way to detect many-body effects in monitored quantum systems without expensive postselection.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Thermodynamic-limit claim rests on finite-size λ0 scaling that is not backed by analytic derivation or larger-L data","rationale":"The reader's verdict (CONDITIONAL) already captures much of this: the rationale explicitly lists 'the thermodynamic-limit claim is an extrapolation from small systems' as a reason for the condition. However, the reader's formal 'weakest_assumption' field identifies diagonalizability/uniqueness of λ0 (exceptional points) as the key assumption. In my reading, the diagonalizability issue is less dangerous: if an exceptional point occurs, the spectral decomposition (35) would need a Jordan-block treatment, but the leading exponential decay rate is still set by Re λ0, so the qualitative claim W_half(τ) ~ e^{λ0τ} survives. By contrast, if λ0's O(1) scaling in strong measurement is a finite-size crossover, the main thermodynamic-limit prediction fails directly. Thus the finite-size extrapolation is the more load-bearing concern. I therefore partially agree with the reader: the concern is present in their rationale but not chosen as the weakest assumption. My proposed test—computing λ0 for L=16/18—would settle the issue. I do not recommend changing the verdict: the paper is technically sound for finite L, but the thermodynamic-limit claim needs stronger numerical support or an analytic argument, which is consistent with CONDITIONAL acceptance.","tokens_in":17417,"tokens_out":9585,"duration_ms":90513,"concrete_test":"Compute λ0 for L=16 and, if feasible, L=18 at γ=1 using a shift-invert sparse eigensolver on the ladder representation (23), exploiting the conserved total particle number of L0, and compare with the L≤14 data in Fig. 2(c). If λ0 remains flat within ~20% of its L=14 value, the O(1) scaling and the thermodynamic-limit claim are supported; if λ0 shows a clear upward trend with L, the O(1) branch of Eq. (1) is a finite-size artifact. An independent cross-check is to extract the long-time slope of the half-chain WTD from quantum-trajectory simulations at L=16, γ=1 with ≥10^7 trajectories, matching it against the computed λ0.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central prediction—Eq. (1), abstract, and Sec. 4.2—that for γ=O(1) the half-chain WTD retains an O(1) decay rate in the thermodynamic limit relies entirely on the observed system-size independence of λ0 in Fig. 2(c). The numerical data there are limited to small systems (the spectra in Fig. 2 are for L=6, and the WTD results go up to L=14 in Appendix B). No analytic scaling argument for why λ0 becomes L-independent at strong measurement is provided; the paper only states the numerical observation. If the apparent flatness of λ0 versus L is a finite-size crossover—for instance, if λ0 eventually grows ∝L beyond the accessible sizes—the anomalous tail would not persist in the thermodynamic limit, and the main claim collapses even though the finite-L spectral framework (Sec. 3.2, Eq. (50)) is internally consistent. This is a missing-support problem rather than an internal inconsistency, but it is load-bearing because the O(1) branch is the paper's headline result. The paper's own limitation paragraph in Sec. 5 mentions exceptional points and subleading eigenvalues, but does not flag the finite-size extrapolation of λ0 scaling.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies waiting-time distributions (WTDs) of quantum jumps in a continuously monitored hard-core boson chain with particle-number detection, whose unconditional steady state is the infinite-temperature state. The authors define a superoperator L0 obtained by removing the jump terms in a subsystem M from the full Liouvillian, and show that the half-chain WTD has a long-time tail governed by the eigenvalue λ0 of L0 with the largest real part, in contrast to the Poissonian WTD of the whole system. Based on exact diagonalization of L0 for small system sizes, they report a qualitative change in the system-size scaling of λ0: for weak measurement λ0 ∝ L, while for strong measurement λ0 is independent of L, implying that the anomalous tail persists in the thermodynamic limit. The WTD is extracted from the spacetime record of jumps and thus does not require postselection.","tokens_in":17758,"tokens_out":10618,"duration_ms":107133,"significance":"If the thermodynamic-limit claim holds, the paper provides a novel, experimentally accessible observable—subsystem waiting-time statistics—that detects many-body effects in monitored quantum dynamics without postselection. The spectral framework connecting the WTD tail to L0 is natural and the exact derivation of the whole-system Poissonian is a clear strength. The numerical simulations (L up to 14) and the explicit comparison of the tail slope with λ0 provide initial support. The paper is also careful to state its main diagonalizability assumption. However, the headline prediction—the O(1) decay rate for strong measurement in the thermodynamic limit—rests on finite-size numerics without an analytic scaling argument or larger-system data, which limits the confidence in the central claim.","major_comments":[{"comment":"The claim that for γ=O(1) the half-chain WTD has a tail ∼e^{−O(1)τ} in the thermodynamic limit is supported only by exact diagonalization of L0 for small systems (Fig. 2 shows spectra for L=6; Appendix B presents WTDs only up to L=14). No analytic scaling argument is given for why λ0 becomes L-independent at strong measurement, and the finite-size data in Fig. 2(c) do not include error bars or a systematic extrapolation. If the apparent constancy of λ0 is a finite-size crossover and λ0 eventually grows with L beyond the accessible sizes, the central prediction of Eq. (1) would not survive. The authors should either provide an analytic argument (e.g., a strong-measurement effective theory showing a finite gap) or present significantly larger-L data using sparse or tensor-network methods.","section":"§4.2, Eq. (1), Fig. 2(c)"},{"comment":"The spectral decomposition in Eq. (50), which yields the tail W_half(τ)∼e^{λ0τ}, explicitly assumes that L0 is diagonalizable and that the eigenvalue λ0 with the largest real part is unique. The paper acknowledges the possible emergence of exceptional points in Sec. 5, but does not verify that these do not occur in the parameter regimes used for the central claim. If exceptional points are present, Eq. (50) must be replaced by a Jordan-form expression, and the tail would acquire a polynomial prefactor. The authors should either check the absence of such degeneracies for the studied γ and L, or clearly state in the main text that the result is conditional on this assumption and discuss the expected modification.","section":"§3.2, Eq. (30), §5"}],"minor_comments":[{"comment":"The normalization proof assumes p_no(∞)=0, but this is only established later in §3.2 via the negativity of λ0. The logical ordering should be adjusted or this decay should be stated as an assumption at this point.","section":"§2.2, Eq. (13)"},{"comment":"The approximation leading to the short-time Poissonian decay is heuristic: 'the local operator n_i can be regarded as approximately conserved' and the replacement e^{L0τ}∼e^{Lτ}e^{−∑_{i∈M}L_iτ} are not rigorously justified. It is acceptable as a physical argument, but the limitations should be stated.","section":"§4.2, Eq. (51)"},{"comment":"Typo: 'Poisonian' should be 'Poissonian' (also in §4.2 and the figure captions). In §2.1, 'trance preserving' should be 'trace preserving'.","section":"§4.1"},{"comment":"For L=14 the number of trajectories is 10^7, and the long-time tail at γ=0.05 appears to be sampled over a narrow time window. Showing error bars or a comparison of the tail over a wider dynamic range would strengthen the claim that the data are consistent with the λ0 slope.","section":"Appendix B"}],"recommendation":"major_revision","confidential_remarks":"The paper contains a clear and promising idea, and the core spectral machinery appears correct. The main risk is the extrapolation from small-system exact diagonalization to the thermodynamic limit in the strong-measurement branch; this is a missing-support issue rather than an internal inconsistency. The authors should be encouraged to provide more substantial evidence for the O(1) scaling of λ0—either an analytic argument or considerably larger system sizes—before publication. The second major issue, the diagonalizability assumption, is acknowledged in the manuscript but should be treated as a caveat in the main text rather than only an outlook item."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The new result here is that the half-chain waiting-time distribution in a continuously monitored many-body system has an anomalous tail controlled by the rightmost eigenvalue λ0 of the no-jump superoperator L0. The whole-system WTD is exactly Poissonian, and the paper shows the half-chain one is not, with the tail slope matching λ0. The derivation of the WTD from the conditional dynamics (Eqs. 6–8) is clean, the spectral representation (50) is correct under the stated assumptions, and the numerical verification in Figs. 3–4 is visually solid. The claim that λ0 is the tail-decay rate is new and reusable beyond this model.\n\nThe soft spot is the thermodynamic-limit statement. Eq. (1) says for γ=O(1), λ0 = -O(1), so the anomalous tail persists as L→∞. The evidence is the system-size scaling in Fig. 2(c), which goes up to L=14. The WTD simulations in Appendix B also stop at L=14. That is a small range for an extrapolation, and no analytic scaling argument is given. It's not an internal inconsistency—the spectral framework itself doesn't depend on the scaling—but the headline claim does. If λ0 eventually grows ∝L beyond accessible sizes, the tail would decay on the same time scale as the Poissonian part, and the 'persistence in the thermodynamic limit' would fail. My guess is the claim is right: in the strong-measurement Zeno limit, the dark state of L0 is a domain wall with only two hopping channels, giving a second-order shift O(J^2/γ) independent of L. But the paper should say something like that or show larger-L λ0 from an iterative solver.\n\nLesser caveats: the diagonalizability and unique-λ0 assumption in Eq. (30) is unchecked; exceptional points are acknowledged but not excluded. This would only turn the pure exponential into a polynomial times exponential, so it's minor. No error bars on the histogram tails, and no code/data, but the qualitative match with λ0 is good.\n\nThis is for people working on monitored quantum dynamics and jump statistics. It deserves peer review; the right call is to send it out with a request to firm up the thermodynamic-limit argument, not to reject.","headline":"New subsystem WTD result, real but needs a sharper thermodynamic-limit argument.","tokens_in":18152,"tokens_out":12806,"would_cite":true,"duration_ms":114995,"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":"The half-chain waiting-time distribution of quantum jumps is not the trivial Poissonian one: its long-time tail is controlled by a single eigenvalue of the no-jump superoperator, and under strong measurement that tail survives as the system","keywords":["waiting-time distribution","quantum jumps","continuous monitoring","Liouvillian spectrum","subsystem statistics","measurement-induced dynamics","postselection-free","hard-core bosons"],"falsifier":"Measure or compute the half-chain WTD at strong measurement γ = O(1) for increasing system sizes (e.g., L = 8, 10, 12, 14) and extract the long-time decay rate. If the rate approaches a nonzero constant as L grows, the persistence claim holds; if the rate decays like 1/L or the tail becomes non-exponential, the λ0-dominated picture fails. An exact-diagonalization scan of L0's spectrum to locate exceptional points in this parameter regime would also settle the question.","tokens_in":17364,"feed_emoji":"⏱️","tokens_out":4022,"duration_ms":41760,"temperature":0.7,"pith_summary":"This paper studies the statistics of waiting times between quantum jumps in a continuously monitored quantum many-body chain whose unconditional steady state is the maximally mixed, infinite-temperature state. While the whole-system waiting-time distribution is exactly Poissonian, the distribution for a half-chain subsystem develops an anomalously heavy tail. The authors trace this tail to the eigenvalue λ0 with the largest real part of the superoperator L0, defined by deleting the subsystem jump terms from the full Liouvillian. They show that λ0 scales with system size for weak measurements but saturates to a size-independent value for strong measurements, implying the anomalous tail persists in the thermodynamic limit only in the strong-measurement regime. Because the distribution is extracted directly from the spacetime record of jumps, it offers a postselection-free, experimentally accessible probe of many-body monitored dynamics.","feed_headline":"Half-chain jump waits keep a long tail as systems grow","feed_subtitle":"A single superoperator eigenvalue rules the anomalous waiting-time tail, giving a postselection-free probe of monitored many-body dynamics.","key_machinery":"The central object is the superoperator L0 = L − Σ_{i∈M} Li, where L is the Lindblad Liouvillian and Li(ρ) = Li ρ Li† are the jump terms in the subsystem M; physically it describes evolution conditioned on no jumps in M. Its right-most eigenvalue λ0, assumed unique and with all real parts ordered as in Eq. (30), dominates the long-time WTD through the spectral decomposition e^{L0 t} = Σ e^{λα t}|ρRα)(ρLα|. The identity λ0 Tr[ρR0] = −Σ_{i∈M} Tr[Li† Li ρR0] proves λ0 < 0 and gives a direct link between the eigenvalue and the decay rate of the no-jump probability.","core_discovery":"The central claim is that the half-chain WTD obeys Whalf(τ) ∼ e^{λ0τ}, where λ0 < 0 is the rightmost eigenvalue of L0 = L − Σ_{i∈M} Li, the superoperator that generates time evolution with no jumps in the half chain while allowing jumps elsewhere. Unlike the full Liouvillian, L0 has no zero eigenvalue, and its spectral decomposition gives the long-time decay of the WTD. The eigenvalue obeys λ0 Tr[ρR0] = −Σ_{i∈M} Tr[Li†Li ρR0], which fixes λ0 < 0 under a mild condition. Numerically, λ0 scales as −O(L) for γ ≪ 1 and as −O(1) for γ = O(1), so the anomalous tail remains robust in the thermodynamic limit only for strong measurement. At short times the WTD still shows the Poissonian slope γL/4 bef","pith_inferences":["A natural extension the paper leaves implicit is to vary the subsystem fraction M/L; the same L0 formalism should produce an M-dependent tail rate, and plotting λ0 against subsystem fraction could expose where the Poissonian whole-chain behavior crosses into the anomalous regime.","The sudden change in λ0 scaling with γ resembles a nonequilibrium crossover; a systematic scaling collapse of λ0/L versus γ, or a derivative of λ0, could locate the crossover measurement strength and test whether it sharpens with system size.","The exceptional points mentioned in Sec. 5 could alter the exponential tail into polynomial corrections; a direct numerical search for eigenvalue coalescence in L0 in the reported parameter regime would show whether the e^{λ0τ} prediction is exact or only approximate.","Because the WTD is extracted from jump records, the same analysis could be applied to other jump observables like factorial cumulants of subsystem jump counts, potentially yielding a family of postselection-free probes."],"forward_implications":["The half-chain WTD is a new, postselection-free observable that carries nontrivial many-body information even when the unconditional steady state is featureless.","For weak measurement the anomalous tail is suppressed exponentially with system size, while for strong measurement the tail rate saturates, so the crossover in λ0 scaling acts as a sharp diagnostic of monitored dynamics.","The whole-system WTD remains exactly Poissonian, so a comparison between full-chain and half-chain waiting times directly reveals the subsystem effect.","The framework gives a spectral route to computing WTDs in other monitored models by diagonalizing the corresponding L0 operator.","The short-time Poissonian slope followed by a λ0-dominated tail means the full statistics are not single-exponential, which is relevant for interpreting continuous-monitoring experiments."],"fun_headline_variants":["One eigenvalue sets the half-chain's anomalous wait tail","Half-chain jump waits persist only for strong monitoring","Postselection-free probe: half-chain wait tail is not Poisson","A single superoperator eigenvalue rules the long wait tail"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The central claim relies on L0 being diagonalizable with a unique eigenvalue λ0 of largest real part (Eq. 30); the paper states this assumption explicitly and notes that exceptional points, where eigenvalues coalesce, could invalidate the pure exponential tail but does not rule them out.","fun_headline_variants_meta":{"raw":{"variants":["One eigenvalue sets the half-chain's anomalous wait tail","Half-chain jump waits persist only for strong monitoring","Postselection-free probe: half-chain wait tail is not Poisson","A single superoperator eigenvalue rules the long wait tail"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001618,"raw_usage":{"total_tokens":6326,"prompt_tokens":842,"completion_tokens":5484,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":586,"completion_tokens_details":{"reasoning_tokens":5420}},"tokens_in":586,"tokens_out":5484,"duration_ms":38603,"temperature":1.0,"reasoning_tokens":5420,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-02T16:58:25.959035+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure or compute the half-chain WTD at strong measurement γ = O(1) for increasing system sizes (e.g., L = 8, 10, 12, 14) and extract the long-time decay rate. If the rate approaches a nonzero constant as L grows, the persistence claim holds; if the rate decays like 1/L or the tail becomes non-exponential, the λ0-dominated picture fails. An exact-diagonalization scan of L0's spectrum to locate exceptional points in this parameter regime would also settle the question.","supporting_citations":[],"review_version":1}