{"id":"c9cc9c61-58b2-4989-81d7-7c9b177838d7","arxiv_id":"2510.23706","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"high","formal_verification":"none","parameter_count":1,"one_line_summary":"It conjectures that the measurement-induced volume-to-area-law entanglement transition for interacting class-DIII Majorana fermions is identical to the noninteracting transition, with the interaction-induced mass being dangerously irrelevant.","lead":"This paper argues that adding fermion interactions does not change the entanglement transition in a monitored one-dimensional Majorana chain: the interacting transition is the same as the noninteracting one. The key mechanism is a 'dangerously irrelevant' interaction term that vanishes at the critical point yet enables volume-law entanglement away from it.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Uncontrolled WZNW dualization: away from R=2 the actual sigma model has no WZNW term, and the dangerously-irrelevant mass conclusion depends on a fitted level parameter x.","rationale":"The reader's weakest assumption identifies the epsilon expansion and analytic continuation to R=1 as the key unverified step. I agree that this is the soft spot, but I would sharpen it: the WZNW dualization is not just 'not fully controlled' - it changes the theory away from R=2. Eq. (1) has no WZNW term for generic R, whereas Eqs. (5) are derived from the WZNW model SO(R)_q. Therefore the continuation from R=2 to R=1 is an ansatz about a different theory, not a controlled perturbative expansion of the original sigma model. The fitted parameter x is load-bearing because the sign of the mass eigenvalue, and hence whether the mass is dangerously irrelevant, depends on x>5/2. The paper's numerical proposals are sensible and would settle the conjecture, but until such a test is performed the central claim remains an unverified conjecture. This does not require changing the reader's CONDITIONAL verdict; it reinforces the need for the proposed numerics. I give partial agreement because the reader noted the lack of control but did not explicitly flag the absence of the WZNW term in the actual theory or the role of x in determining the sign of the mass eigenvalue.","tokens_in":18465,"tokens_out":10898,"duration_ms":115683,"concrete_test":"Simulate the 1D monitored interacting Majorana class-DIII circuit with weak local density-density interactions U>0 (e.g., using matrix product state time evolution with measurements, or a hybrid fermion-Gaussian/perturbative interaction scheme) at the noninteracting critical measurement rate. Extract the effective central charge c_eff from the Shannon entropy of the measurement record on cylinders and/or the entanglement central charge c_ent from subsystem entanglement entropies, and compare with the noninteracting DIII values reported in Ref. [34]. If c_eff or c_ent differ from the noninteracting values beyond statistical error, the MIPT universality class is not the noninteracting one, directly refuting the dangerously irrelevant mass conjecture. If they match, the concern about the uncontrolled epsilon expansion is substantially mitigated.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that the interacting DIII MIPT is identical to the noninteracting one because the interaction-induced mass is dangerously irrelevant. The evidence is the R=2-epsilon RG flow Eq. (5), obtained by non-abelian bosonization in SO(R)_q with q=8-x*epsilon. The authors explicitly state this is 'a technical trick (dualization)' and that Eq. (1) 'does not possess a WZNW term' for generic R. Thus the beta functions used to locate the fixed point and to compute the mass eigenvalue are not the beta functions of the derived field theory except at R=2. Moreover, the mass eigenvalue at the fixed point is (epsilon/4)(5-2x); the dangerously irrelevant scenario requires x>5/2. The parameter x is not fixed by the microscopic model; it is set by matching the noninteracting correlation-length exponent nu~2.1. If a more direct calculation or an independent determination gave x<=5/2, the mass would be relevant and the interacting MIPT would be a genuinely new universality class. The paper's own caveat ('not fully controlled') plus the fitted x means the central claim is a conjecture with analytic support, not an established result.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the measurement-induced entanglement transition (MIPT) in one-dimensional monitored, interacting Majorana fermions with no conserved quantities (class DIII). It formulates a replicated Keldysh nonlinear sigma model with an SO(R) target manifold and an interaction-induced replica-anisotropic 'mass' term OM. The paper conjectures that the interacting DIII MIPT is in the same universality class as the noninteracting DIII MIPT, because the mass term is a dangerously irrelevant perturbation at the critical point and only becomes relevant away from criticality, generating the volume-law phase. As analytic support, the authors use a non-abelian bosonization (SO(R)_q WZNW) description and an R=2−ε expansion, identifying a noninteracting fixed point and computing RG eigenvalues with a level-deformation parameter x. Setting x=3 reproduces the known noninteracting correlation-length exponent ν≈2.1. The paper also proposes several concrete numerical tests (effective central charge, entanglement central charge, fermion-correlation exponents, reduced-density-matrix level statistics) to test the conjecture. The manuscript is explicit that the central claim is a conjecture and that the epsilon expansion is 'not fully controlled.'","tokens_in":18836,"tokens_out":5267,"duration_ms":58550,"significance":"If correct, the conjecture is significant: it would show that interactions can leave the MIPT universality class unchanged even though they are essential for the existence of a true volume-law phase, and it would distinguish class DIII from class AIII where interactions are known to modify the transition. The paper gives credit for proposing falsifiable numerical tests and for being transparent about the status of the central claim. However, the analytic evidence is not a derivation from the microscopic model: the dangerous-irrelevance conclusion rests on an uncontrolled analytic continuation in the replica number and on a free parameter x that is fixed by matching the very noninteracting physics whose stability is being asserted. The numerical tests, if carried out, would be a genuine test of the conjecture, but they are not part of the present manuscript.","major_comments":[{"comment":"The central claim — that the mass operator is dangerously irrelevant at the interacting DIII MIPT — is extracted from the one-loop beta functions (5), obtained by writing the R=2 theory as an SO(R)_q WZNW model and continuing to R=2−ε. The authors explicitly state that Eq. (1) has no WZNW term for generic R and that the WZNW encoding is 'a technical trick (dualization)' (Introduction). Thus the beta functions used to locate the fixed point (15) and to compute the mass eigenvalue are not beta functions of the actual sigma model away from R=2. The dangerous-irrelevance conclusion is therefore an extrapolation, not a derivation from the microscopic model. This is load-bearing evidence for the conjecture, so the manuscript should either provide an independent check (e.g., a direct RG calculation in the sigma model, or a numerical test that constrains x) or state prominently that the result i","section":"Epsilon expansion, Eq. (5) and Appendix C, Eq. (14)"},{"comment":"The parameter x is a free input, not fixed by the microscopic Hamiltonian. It is set to x=3 by matching the noninteracting correlation-length exponent ν≈2.1. The mass eigenvalue at the noninteracting fixed point is (ε/4)(5−2x), so dangerous irrelevance requires x>5/2. The fitted x=3 satisfies this, but the conclusion is therefore contingent on a fit to the very noninteracting universality class whose stability is being asserted. If an independent calculation or numerical simulation gave x≤5/2, the mass would be relevant and the interacting MIPT would belong to a new universality class. This fitted-parameter dependence should be quantified or removed; at minimum, the paper should explicitly flag that the entire 'same universality class' claim hinges on the value of x.","section":"Epsilon expansion, text after Eq. (5) and Eq. (15)"},{"comment":"The physical argument for the dangerous irrelevance of the mass relies on the assumption that the entanglement-generation rate vanishes continuously at a continuous MIPT. This is plausible but is an assumption, not a consequence of the RG calculation. If the transition is continuous only in some coarse-grained sense while the local rate remains finite at criticality, the mass could be relevant and the interacting MIPT could differ from the noninteracting one. The paper should explicitly separate this physical assumption from the field-theoretic calculation, and note that the proposed numerical tests are also needed to test this assumption directly.","section":"Field theory and nature of the MIPT"}],"minor_comments":[{"comment":"The title has a broken space: 'Entanglement T ransition' should be 'Entanglement Transition'.","section":"Title"},{"comment":"The sentence 'We conjecture that x > 2/5' appears to contain a typo. The fixed-point analysis requires 5/2 < x < 4, so the conjecture should read x > 5/2. Please check all occurrences of this inequality.","section":"Epsilon expansion, text near Eq. (5)"},{"comment":"The caption states that the mass parameter yM is 'an irrelevant perturbation' in subpanel (a). It would be clearer to say 'dangerously irrelevant' and to explain that the relevance of yM away from the critical surface is what generates the volume-law phase.","section":"Figure 2 caption"},{"comment":"The argument that the fermion-bilinear correlation exponent is exactly 2 for the noninteracting circuit relies on the Noether-current argument at the SO(R)×SO(R) symmetric fixed point. It would help to state explicitly that this argument does not apply to the interacting case, which is precisely why it is a useful diagnostic.","section":"Appendix B"}],"recommendation":"major_revision","confidential_remarks":"The paper is honest about the conjectural status of its central claim, and the proposed numerical tests are concrete and valuable. My main concern is that the analytic support depends on the fitted parameter x and on an uncontrolled WZNW dualization; if the authors can either provide an independent constraint on x or clearly reframe the result as a highly-motivated conjecture with the x-dependence made explicit, the paper could be suitable for publication. I would not recommend rejection because the manuscript does not overclaim its status and the numerical program is a legitimate route to resolving the issue."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nIf you only remember one thing: this is a conjecture paper, not a proof paper. The authors argue that the monitored interacting Majorana chain in class DIII has the same MIPT as the noninteracting one, with interactions generating a dangerously irrelevant mass. The evidence is a one-loop RG flow obtained in an R=2-ε expansion, and they say themselves that the expansion is \"not fully controlled.\" I'd send it to a referee, but I would not treat the conjecture as established.\n\nWhat's genuinely new: the conjecture itself (interacting DIII = noninteracting DIII) and the epsilon-expansion calculation around R=2, extending the Fu-Kane trick to the interaction-deformed problem. The physical picture is clean: the mass operator is an interparticle scattering rate density, and if it is proportional to the entangling rate density, it must vanish continuously at a continuous MIPT. That argument is worth the paper. The proposed numerical tests (ceff, cent, correlation-function exponents) are concrete and could distinguish the interacting from noninteracting universality class.\n\nThe soft spot, in one paragraph: the beta functions in Eq. (5) come from an SO(R)_q WZNW model with q=8-xε. The actual field theory, Eq. (1), has no WZNW term for generic R; the WZNW encoding is explicitly a \"technical trick\" to capture R=2 physics. So the flow used to locate the fixed point and compute the mass eigenvalue (ε/4)(5-2x) is not the beta function of the derived theory except at R=2. And x is not fixed by the microscopic model; it is chosen to reproduce the noninteracting correlation-length exponent ν≈2.1. The dangerously irrelevant scenario requires x>5/2, and the chosen x=3 satisfies it. If a more direct calculation gave x≤5/2, the mass would be relevant and the interacting MIPT would be a genuinely different universality class. The paper is honest about this; the caveat and the fitted parameter are exactly where the claim is weakest.\n\nBottom line: the conjecture is plausible, the RG evidence is suggestive but not controlled, and the authors propose the right numerical experiments to settle it. Worth serious peer review. I'd cite it for the conjecture and the physical picture.\n\nBest,\n[You]","headline":"A well-posed conjecture with honest caveats: the epsilon-expansion evidence is real but uncontrolled, and the dangerously-irrelevant conclusion rests on a fitted parameter.","tokens_in":19191,"tokens_out":2998,"would_cite":true,"duration_ms":28461,"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":"Interactions do not change the universality class of the Majorana measurement-induced transition: the interaction-induced mass is dangerously irrelevant at the critical point.","keywords":["measurement-induced phase transition","entanglement transition","monitored Majorana fermions","class DIII","replica nonlinear sigma model","dangerously irrelevant mass","R=2−ε expansion","volume-law phase"],"falsifier":"A numerical simulation of the interacting class-DIII monitored circuit could measure the entanglement central charge at the transition: if c_ent differs from the noninteracting value 0.39 ± 0.02 reported in the paper by more than the statistical error, or if the correlation-length exponent ν departs significantly from ≈ 2.1, then the conjecture that interactions do not alter the critical point would be falsified.","tokens_in":18419,"feed_emoji":"⚛️","tokens_out":6839,"duration_ms":65461,"temperature":0.7,"pith_summary":"What happens to the volume-law–to–area-law entanglement transition of a monitored fermion chain when interactions are switched on? This paper argues that for one-dimensional, monitored, interacting Majorana fermions with no conserved quantities (symmetry class DIII), the transition is controlled by exactly the same critical theory as the noninteracting system. The interaction-generated 'mass' term in the field theory — which encodes interparticle scattering and drives volume-law entanglement — becomes dangerously irrelevant at the critical point: it flows to zero exactly at the transition, restoring the free-fermion critical point, while any deviation from the critical surface sends the system into the volume-law phase. The argument uses a replicated Keldysh nonlinear sigma model and a controlled ε-expansion in the replica number, and it yields concrete numerical predictions that would confirm or refute the conjecture.","feed_headline":"Interactions don't move the Majorana entanglement transition","feed_subtitle":"The interaction-induced mass is dangerously irrelevant, leaving free-fermion criticality intact at the transition.","key_machinery":"The central object is the replicated Keldysh nonlinear sigma model (a field theory whose field X is an SO(R) rotation matrix), with action S = (λ/16) tr(∇X^T·∇X) − (M/16) O_M, where λ is the stiffness (inversely proportional to the measurement rate) and the 'mass' term O_M = ∑_{j,k} X_{jk}^4 represents the interaction-induced scattering (entangling) rate. The mass breaks the continuous replica symmetry SO(R)×SO(R) down to discrete permutations S_R×S_R. The analysis proceeds through a technical duality to the SO(R)_q Wess-Zumino-Novikov-Witten model at level q = 8, which allows an ε-expansion in the replica number R = 2 − ε; the resulting one-loop RG flow shows the mass to be dangerously irre","core_discovery":"The central claim is that the measurement-induced phase transition (MIPT) in a one-dimensional chain of monitored, interacting Majorana fermions (symmetry class DIII, no conserved quantities) has exactly the same critical point as the noninteracting transition. The interaction-induced replica-anisotropic 'mass' operator O_M = ∑_{j,k} X_{jk}^4 — which represents a local interparticle scattering rate, i.e., the entangling rate density — becomes dangerously irrelevant at the critical fixed point. At the transition the mass flows to zero and the noninteracting SO(R)×SO(R)-symmetric critical theory (a non-unitary conformal field theory) is recovered; a nonzero mass deviation instead drives the sy","pith_inferences":["This suggests a broader principle: in monitored fermion systems without continuous symmetries, the dangerously-irrelevant-mass mechanism could protect free-fermion criticality, potentially unifying MIPT universality classes across interacting circuits.","The Fermi-golden-rule logic — that a continuous transition forces the entangling rate to vanish — implies that the mass scaling dimension at any continuous MIPT must exceed 2; a future example with a relevant mass would challenge the picture.","One could test the dangerously-irrelevant scenario directly by measuring the time-decay of the fermion-bilinear correlation function as the critical point is approached from the volume-law side; the mass-induced decay rate should vanish continuously at the transition.","The ε-expansion predicts that the effective central charge of the interacting transition equals that of the noninteracting one; a numerical measurement differing by more than a few percent would constitute evidence against the conjecture."],"forward_implications":["The interacting and noninteracting DIII MIPTs share the same critical exponents and effective (central) charges; this can be tested by comparing c_eff and c_ent in numerical simulations of the two circuits.","At the critical point, interactions flow to zero and an emergent continuous replica symmetry appears, so free-fermion techniques can be used to compute universal properties of the interacting transition.","A nonzero interaction strength only matters away from the critical point, where it induces the volume-law phase; the area-law phase is unaffected by interactions at sufficiently low measurement rates.","The correlation-length exponent satisfies ν ≈ 2.1 (for x = 3), consistent with the Chayes–Harris bound; the same exponent should govern the interacting transition.","For monitored fermions with extra continuous symmetries (e.g., conserved U(1) charge), interaction-induced terms from Noether currents are not dangerously irrelevant and the critical theory is expected to differ."],"fun_headline_variants":["Interactions don't move the Majorana MIPT","Majorana entanglement transition unchanged by interactions","Interactions prove dangerously irrelevant at Majorana MIPT","Free-fermion criticality survives interactions in monitored Majoranas","Interaction mass vanishes at Majorana measurement transition"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The entire argument rests on the ε-expansion in the replica number R = 2 − ε, combined with non-abelian bosonization at level q = 8 and analytic continuation to R = 1, correctly describing the strong-coupling physical fixed point; the authors explicitly state this approach is 'not fully controlled.'","fun_headline_variants_meta":{"raw":{"variants":["Interactions don't move the Majorana MIPT","Majorana entanglement transition unchanged by interactions","Interactions prove dangerously irrelevant at Majorana MIPT","Free-fermion criticality survives interactions in monitored Majoranas","Interaction mass vanishes at Majorana measurement transition"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000193,"raw_usage":{"total_tokens":1214,"prompt_tokens":797,"completion_tokens":417,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":541,"completion_tokens_details":{"reasoning_tokens":344}},"tokens_in":541,"tokens_out":417,"duration_ms":4285,"temperature":1.0,"reasoning_tokens":344,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-04T07:48:50.309065+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A numerical simulation of the interacting class-DIII monitored circuit could measure the entanglement central charge at the transition: if c_ent differs from the noninteracting value 0.39 ± 0.02 reported in the paper by more than the statistical error, or if the correlation-length exponent ν departs significantly from ≈ 2.1, then the conjecture that interactions do not alter the critical point would be falsified.","supporting_citations":[],"review_version":1}