{"id":"7f4799a4-b19a-4148-b2e7-5d8dfc77dfe5","arxiv_id":"2502.02001","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"In the non-Hermitian five-state Potts model, complex entanglement entropy built from biorthogonal reduced density matrices yields a complex central charge matching the complex CFT prediction.","lead":"Complex entanglement entropy, computed from the right and left ground states of a non-Hermitian Hamiltonian, follows the conformal field theory scaling predicted for a complex central charge in the five-state Potts model. The result offers a practical observable for detecting complex conformal field theory in open quantum many-body systems.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The universal claim rests on a three-point PBC fit, while the paper's own OBC data deviate strongly without a quantitative boundary-effect model; a finite-size OBC check is needed to distinguish boundary effects from a breakdown of the biorthogonal-entropy/CFT relation.","rationale":"The reader's conditional verdict is reasonable. The paper has genuine strengths: the DMRG convergence analysis, the agreement with the theoretical central charge in Eq. (9), and the comparison with Hermitian and SVD based entropy definitions all provide independent evidence that the biorthogonal object is the right quantity. I do not see an internal inconsistency in the PBC numerical procedure. However, the load-bearing gap is the generality of the claim: the CFT formula for the biorthogonal entropy is assumed rather than derived for complex central charges, and the only quantitative success is a three-point PBC fit. The paper itself reports an OBC failure, and the explanation via non-Hermitian skin effects is plausible but unquantified. A simple finite-size OBC study would either validate that explanation or reveal that the biorthogonal entropy is boundary-condition dependent in a way not controlled by the CFT. Pending that check, the central claim should not be upgraded to full acceptance, but it should not be rejected either; the existing conditional verdict remains the correct one.","tokens_in":16499,"tokens_out":7389,"duration_ms":83521,"concrete_test":"Repeat the OBC calculation at L=24, 32, and 48 with D≥800 and fit the central three points to Eq. (31); also fit the full curve with a correction term, S(l) = (c/6) log[(2L/pi) sin(pi l/L)] + S0 + a/l + b/(L-l). If Re c and Im c approach 1.1376 − 0.0211i with increasing L, the boundary-effect explanation is supported; if they remain near 1.44 + 0.05i, the biorthogonal entropy does not follow complex CFT under OBC and the universality claim must be restricted, at minimum, to periodic boundary conditions.","verdict_should_be":"UNCHANGED","load_bearing_attack":"In Eq. (30), the paper claims that the biorthogonal entanglement entropy obeys the standard CFT formula with a complex central charge, but no analytic derivation is supplied for this construction in a complex CFT. The numerical support is a single non-Hermitian Potts Hamiltonian under periodic boundary conditions, with c extracted from only the three central values of l (Fig. 1 and Table I). The same protocol under OBC (Eq. (31), Fig. 3) yields Re c ≈ 1.4415 and an imaginary part that does not follow the CFT scaling; the paper attributes this to non-Hermitian boundary effects without a quantitative model. Because non-Hermitian systems are known to be extremely sensitive to boundary conditions, the OBC failure is the most direct threat to the universal claim. The DMRG bond-dimension convergence checks only exclude truncation error; they do not establish that the remaining finite-size corrections are small enough to make the three-point PBC fit an unbiased estimate of c. Thus the central assertion remains conditional on the assumption that the PBC agreement is a bulk CFT effect and that the OBC deviation is a benign finite-size boundary effect.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper introduces a complex-valued entanglement entropy constructed from the biorthogonal (right/left) ground states of non-Hermitian Hamiltonians and applies it to the critical non-Hermitian five-state Potts model, which is believed to be described by complex conformal field theory with central charge c = 1.13755 - 0.02107i. Using finite-system DMRG, the authors compute the entropy as a function of subsystem size for periodic and open boundary conditions. By fitting the universal CFT scaling form to three central points, they extract complex central charges that approach the theoretical value for PBC with increasing system size (e.g., c = 1.1385 - 0.0205i at L=20, D=400). The OBC data, however, yield a significantly different real part (Re c ≈ 1.4415) and a positive imaginary part (Im c ≈ 0.05), which the paper attributes to non-Hermitian boundary effects without quantitative analysis. The paper also compares alternative entropy definitions and examines the Petermann factor to contrast with c = -2 nonunitary CFT.","tokens_in":16766,"tokens_out":7046,"duration_ms":65595,"significance":"If the PBC agreement is taken at face value, the result would be a useful numerical demonstration that biorthogonal entanglement entropy can serve as a probe of complex CFT, complementing previous energy-based extractions of complex central charges. The paper's strengths are that the fitted central charge is compared with an independent theoretical prediction (Eq. 9) from the literature, that bond-dimension convergence is checked for the PBC and OBC data, and that alternative definitions (Sabs, SVD entropy) are analyzed to show the specificity of the biorthogonal construction. However, the central claim of universality is weakened by the unexplained OBC failure, by the lack of error bars and of a full finite-size scaling analysis, and by the absence of any analytic derivation of the assumed CFT form. The paper is therefore best viewed as a promising numerical study that needs additional systematic checks before the universality claim can be considered established.","major_comments":[{"comment":"The fit protocol used throughout the paper is a least-squares fit of Eq. (30) (or (31)) to only the three central values of l for each L, with no reported error bars or residuals. This leaves one degree of freedom (three data points, two fit parameters) and makes the extracted central charge highly sensitive to boundary distortions: Fig. 1(b) shows clear deviations for l near 1 and L-1, and Table I shows that the L=4 system yields Im c = +0.0104, opposite in sign to the CFT value -0.0211. The paper attributes these deviations to boundary effects but provides no quantitative model for them. To support the universality claim, the authors should either fit the full l-dependence with an explicit boundary-correction term or demonstrate that the three-point estimate converges with L in a controlled way, e.g., by a finite-size extrapolation.","section":"IV.B, Eq. (30), Fig. 1, Table I"},{"comment":"The open-boundary results in §IV.C are in sharp conflict with the CFT prediction: the fit gives Re c ≈ 1.4415 and Im c ≈ 0.05, whereas Eq. (9) predicts Re c = 1.1376 and Im c = -0.0211. The authors invoke 'subtle boundary effects' and the non-Hermitian skin effect, but no quantitative boundary-effect analysis is presented. Since Eq. (31) is the standard CFT formula for open boundaries, the observed failure means that the paper's central claim — that the universal scaling S = (c/3) log[...] holds for complex CFT — is not supported by the OBC geometry. The manuscript needs either a demonstration that the OBC deviations vanish as L increases, or a concrete model of the non-Hermitian boundary contributions that reproduces the fitted values; without this, the PBC success cannot be unambiguously separated from boundary artifacts.","section":"IV.C, Eq. (31), Fig. 3"},{"comment":"Equation (30) is posited as the CFT scaling law for the biorthogonal entanglement entropy, but no derivation is given for why the non-Hermitian reduced density matrix (15) should obey this form, nor why the central charge should appear as a complex coefficient in the logarithm. The numerical test is therefore a check that the coefficient extracted from data matches Eq. (9) under the assumption that the l-dependence is log[(L/π) sin(πl/L)]. The paper should explicitly state that this is an assumed functional form rather than a derived result, and ideally should provide a derivation or a more direct test, for example by verifying the predicted l-dependence over the full range after subtracting a modeled boundary term. This issue is load-bearing because the universality claim rests entirely on this assumed form.","section":"III, Eq. (17), and IV.B"}],"minor_comments":[{"comment":"The paper uses 'we demonstrate' in the abstract and introduction, but the evidence is numerical and conditional on the assumed CFT form; a more cautious phrasing such as 'we provide numerical evidence' would better match the content.","section":"Abstract and Section I"},{"comment":"The statement that the L=20, D=300 imaginary part is a coincidence due to competition between finite-size and bond-dimension effects is informal; a quantitative estimate of both corrections would be more persuasive.","section":"IV.B"},{"comment":"The DMRG convergence analysis reports the difference between successive bond dimensions, but not the truncation error or the overlap between right and left ground states; specifying these would strengthen the reliability argument.","section":"IV.A and Figs. 1(c)-(d), 3(c)-(d)"},{"comment":"The comparison with Sabs and Ssvd is presented without a quantitative explanation of why the differences appear where they do; a few sentences on the origin of these deviations would improve the discussion.","section":"IV.D, Fig. 4"}],"recommendation":"major_revision","confidential_remarks":"For the editor: The paper is a solid numerical study, but its central claim is currently supported only by the PBC data. The OBC discrepancy is serious and needs to be addressed before publication. I also recommend requesting error bars for the extracted central charges and a fuller discussion of the assumed CFT scaling form. The comparison with the alternative definitions (Sabs and SVD entropy) is useful but not decisive."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's the short version: the paper shows, numerically, that the biorthogonal entanglement entropy built from right and left ground states reproduces the expected complex central charge of the non-Hermitian five-state Potts model for periodic boundary conditions. That specific test—complex c extracted from entanglement scaling in a lattice model—is new, and the DMRG work is careful enough that I believe the PBC result. But the paper claims more than it shows: the clean agreement comes from fitting only the three central values of l, without error bars, and the same protocol fails badly under open boundary conditions, with Re c ≈ 1.44 instead of ≈ 1.14, and an imaginary part that does not follow the CFT form. The authors attribute the OBC failure to boundary effects and skin-effect sensitivity, but they do not quantify it. That is the load-bearing soft spot.\n\nWhat I like: they do not hide the OBC problem—they present it directly. The comparison with SVD and absolute-value entropies is a useful sanity check, and the Petermann-factor contrast with the c = −2 CFT helps separate their regime from the exceptional-point-dominated one. The writing is clean, the branch-cut convention is explicit, and the citation pattern is appropriate: prior biorthogonal-entropy work, the lattice model of Tang et al., and the pseudo-entropy literature are all credited.\n\nConcerns, in order of severity: (1) The three-point fit is fragile; the L = 4 case gives the wrong sign for Im c, which shows how much finite-size/boundary contamination can shift the extracted value. A few more subsystem sizes, or a fit with residuals, would make the PBC claim sturdier. (2) No analytic argument explains why the Calabrese–Cardy formula with complex c should hold for the biorthogonal entropy; the paper assumes the continuation. That is a reasonable starting point, but it makes the paper a numerical check, not a derivation, and one-model support is weak for the word \"universal.\" (3) No data or code are released, so reproducing the numbers requires reimplementing non-Hermitian DMRG. (4) The OBC deviation is dismissed too quickly; a finite-size scaling study of the boundary region, or a model of how the skin effect contaminates the entropy, would address the most direct challenge to the claim.\n\nThe PBC central argument holds up, and the paper is honest about its limitations. I would send it to a serious referee, but ask for error bars, a more robust fitting protocol, and a quantitative OBC analysis before publication. It is a useful paper for people working on non-Hermitian criticality, complex CFT, and pseudo-entropy; for that audience it deserves careful reading.","headline":"A solid numerical demonstration that biorthogonal entanglement entropy extracts the complex central charge of the non-Hermitian Potts model, but the universal claim is undercut by an unexplained OBC deviation and a fragile three-point fit.","tokens_in":17246,"tokens_out":3080,"would_cite":true,"duration_ms":31457,"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":"Complex entanglement entropy confirms complex CFT scaling","keywords":["complex conformal field theory","entanglement entropy","non-Hermitian quantum systems","five-state Potts model","density matrix renormalization group","complex central charge","pseudo-entropy","nonunitary critical phenomena"],"falsifier":"An exact or high-precision computation for larger systems, say $L = 32$ with bond dimension $D = 600$, that shows the fitted complex central charge moving away from $c \\approx 1.13755 - 0.02107i$, or an imaginary part of $S(L, L/2)$ that does not grow as $\\frac{\\mathrm{Im}\\, c}{3} \\log\\left(\\frac{L}{\\pi}\\right)$, would disprove the claim. The open-boundary result already provides a partial negative control: a robust theory would need to explain why the same fitting protocol gives $\\mathrm{Re}\\, c \\approx 1.4415$ there.","tokens_in":16327,"feed_emoji":"🔗","tokens_out":6332,"duration_ms":56323,"temperature":0.7,"pith_summary":"The paper argues that a complex-valued generalization of entanglement entropy can characterize conformal field theory (CFT) with complex central charges, which describe non-Hermitian critical systems. It constructs non-Hermitian reduced density matrices from the right and left ground states and defines the entropy through the complex logarithm. For the critical non-Hermitian five-state Potts model, density-matrix-renormalization-group calculations yield the universal CFT scaling $S = \\frac{c}{3} \\log\\left(\\frac{L}{\\pi} \\sin\\frac{\\pi l}{L}\\right) + S_0$ with complex $c$. The fitted central charge approaches the CFT prediction $c \\approx 1.13755 - 0.02107i$, and the paper reads this as evidence that complex CFT controls the model's critical behavior.","feed_headline":"Complex entanglement entropy confirms complex CFT scaling","feed_subtitle":"A biorthogonal reduced density matrix makes entropy complex and exposes the universal scaling law.","key_machinery":"The central object is the non-Hermitian (biorthogonal) reduced density matrix $\\hat{\\rho}^{RL}_A = \\mathrm{tr}_B\\left(|\\psi\\rangle\\langle\\langle\\psi| / \\langle\\langle\\psi|\\psi\\rangle\\right)$, built from a right ground state $|\\psi\\rangle$ and a left ground state $|\\psi\\rangle\\rangle$ of a non-Hermitian Hamiltonian. The complex entanglement entropy is $S = -\\mathrm{tr}_A\\left(\\hat{\\rho}^{RL}_A \\log \\hat{\\rho}^{RL}_A\\right)$ with the logarithm branch $\\arg z \\in (-\\pi, \\pi]$. This object carries the argument because it is the quantity whose finite-size scaling reproduces the CFT formula, and it is the ingredient that makes the entropy complex rather than real.","core_discovery":"Conformal field theory's entanglement scaling formula extends without modification to complex central charges, provided the reduced density matrix is built biorthogonally from both right and left eigenstates rather than from a single Hermitian density matrix. The resulting entropy is complex, and its real and imaginary parts both follow the universal finite-size scaling predicted by CFT under periodic boundary conditions. Numerical fitting for the non-Hermitian five-state Potts model gives $c \\approx 1.1385 - 0.0205i$ at $L = 20$, $D = 400$, close to the analytic $c \\approx 1.13755 - 0.02107i$, and closer to theory than the central charge extracted from eigenenergy scaling in earlier work. The authors take this as validation that complex entanglement entropy is a meaningful probe of complex CFT and of open quantum many-body criticality.","pith_inferences":["If the scaling law holds for other $Q > 4$ Potts models or complexified Kondo couplings, complex entanglement entropy would offer a general route to determining complex fixed points from lattice data.","The open-boundary deviation suggests a boundary-condition-dependent effective description, possibly connected to the non-Hermitian skin effect, which could be tested by studying how the fitted $c$ depends on boundary coupling.","A direct extension would be to compute the complex Rényi entropy and check whether its $n$-dependence matches the CFT prediction, giving an independent falsifier beyond von Neumann entropy.","The Petermann factor analysis implies that the finite-size spectral gap at the complex fixed point is what makes the CFT scaling visible; systems with exceptional points may require a different scaling ansatz."],"forward_implications":["The universal formula $S = \\frac{c}{3} \\log\\left(\\frac{L}{\\pi} \\sin\\frac{\\pi l}{L}\\right) + S_0$ can be used to extract complex central charges directly from entanglement data.","Complex entanglement entropy is a numerical probe of complex CFT that is more precise than finite-size eigenenergy scaling, because it avoids an extra energy-scale fit.","The same construction connects to pseudo-entropy in high-energy physics, so CFT results for complex entropy may transfer to that setting.","Real-valued alternatives based on Hermitian reduced density matrices or SVD fail to capture the complex central charge, indicating that biorthogonality is essential.","Open boundary conditions do not obey the same scaling, signaling boundary effects specific to non-Hermitian systems that require separate treatment."],"supporting_citations":[{"why":"Supplies the non-Hermitian five-state Potts model Hamiltonian and its complex fixed point parameters, the system the paper studies.","marker":"[57]"},{"why":"Derives the complex central charge $c = 1.13755 \\pm 0.02107i$ from analytic continuation of the Potts model, the target value for the numerical fit.","marker":"[48]"},{"why":"Gives the CFT entanglement entropy scaling formula used for periodic boundary conditions.","marker":"[9]"},{"why":"Provides the conformal-field-theory scaling formulas, including the open-boundary form, used in the fits.","marker":"[10]"},{"why":"Introduces biorthogonal reduced density matrices and generalized entanglement entropy for nonunitary CFT, the construction the paper extends to complex central charges.","marker":"[62]"},{"why":"Provides a lattice realization of complex CFT for the $Q > 4$ Potts model, linking the lattice model to complex CFT.","marker":"[56]"},{"why":"Describes the DMRG truncation procedure adapted to non-Hermitian systems, which the paper uses to obtain right and left ground states.","marker":"[88]"},{"why":"Supplies the tensor-network software used for the DMRG calculations.","marker":"[90]"}],"fun_headline_variants":["Complex entropy confirms complex CFT scaling","Biorthogonal entropy validates complex CFT","Complex entanglement entropy follows CFT scaling","Non-Hermitian Potts model tests complex entropy"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument assumes that the entropy computed from paired right and left ground-state wave functions obeys the same CFT scaling formula as a Hermitian system's entropy in the middle of the chain, so fitting only the three central subsystem sizes yields an unbiased estimate of the complex central charge.","fun_headline_variants_meta":{"raw":{"variants":["Complex entropy confirms complex CFT scaling","Biorthogonal entropy validates complex CFT","Complex entanglement entropy follows CFT scaling","Non-Hermitian Potts model tests complex entropy"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00014,"raw_usage":{"total_tokens":1099,"prompt_tokens":823,"completion_tokens":276,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":439,"completion_tokens_details":{"reasoning_tokens":220}},"tokens_in":439,"tokens_out":276,"duration_ms":3021,"temperature":1.0,"reasoning_tokens":220,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-09T13:42:41.472805+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"An exact or high-precision computation for larger systems, say $L = 32$ with bond dimension $D = 600$, that shows the fitted complex central charge moving away from $c \\approx 1.13755 - 0.02107i$, or an imaginary part of $S(L, L/2)$ that does not grow as $\\frac{\\mathrm{Im}\\, c}{3} \\log\\left(\\frac{L}{\\pi}\\right)$, would disprove the claim. The open-boundary result already provides a partial negative control: a robust theory would need to explain why the same fitting protocol gives $\\mathrm{Re}\\, c \\approx 1.4415$ there.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the non-Hermitian five-state Potts model Hamiltonian and its complex fixed point parameters, the system the paper studies."},{"cited_title":"Gorbenko, S","cited_arxiv_id":null,"evidence_quote":"Derives the complex central charge $c = 1.13755 \\pm 0.02107i$ from analytic continuation of the Potts model, the target value for the numerical fit."},{"cited_title":"Chang, J.-S","cited_arxiv_id":null,"evidence_quote":"Introduces biorthogonal reduced density matrices and generalized entanglement entropy for nonunitary CFT, the construction the paper extends to complex central charges."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides a lattice realization of complex CFT for the $Q > 4$ Potts model, linking the lattice model to complex CFT."},{"cited_title":"Carlon, M","cited_arxiv_id":null,"evidence_quote":"Describes the DMRG truncation procedure adapted to non-Hermitian systems, which the paper uses to obtain right and left ground states."}],"review_version":1}