{"id":"91532192-93c5-45a4-8885-abd9398ab8ac","arxiv_id":"2608.01680","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"In a PT-symmetric 1D moiré lattice, the parity of the commensurate ratio denominator determines which band pair first breaks PT symmetry, controlling whether the lowest flat band broadens monotonically (even denominators) or nonmonotonically (odd denominators).","lead":"The paper studies a 1D moiré optical lattice with PT-symmetric gain/loss and shows that whether the lattice-period ratio has an even or odd denominator controls how the flat lowest band responds to dissipation: even denominators monotonically broaden the band, odd denominators produce a nonmonotonic response, and weak interactions modify these trends in a parity-dependent way.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Odd-q parity rule is proven only for q=3, rests on an unproved center-dominance assumption for q≥5, and is literally contradicted by q=1.","rationale":"The central claim is that the parity of q controls the PT-breaking sequence and the monotonic versus nonmonotonic flatness response in the model of Eq. (6). For this to hold for every odd q, two conditions must be true: (i) the lowest odd-q eigenstate is dominated by the central self-conjugate minimum, and (ii) the first PT-breaking pair among the remaining bands is (2,3). The paper proves both for q=3, checks q=5 numerically, and gives a large-γ fit for q=7. The general-q tight-binding block structure in Appendix B3 shows the center site couples only indirectly to the imaginary potential, but it does not show that the center site dominates the lowest eigenstate for all odd q or that the resulting first PT-breaking pair is always (2,3). Appendix B4 shows the same geometric center argument can fail when phases are changed, so the Class-I restriction is essential. In addition, q=1 in Fig. 2(a) is an actual counterexample to the unqualified abstract statement. None of this invalidates the specific q=3,5 results or the even-q branch, so the paper remains conditionally acceptable with a scope qualification and a numerical check for larger odd q. This matches the reader's conditional verdict and weakest assumption, so no verdict change is needed.","tokens_in":21038,"tokens_out":9963,"duration_ms":98311,"concrete_test":"Perform continuum plane-wave diagonalization (Appendix A, d=10q) for the noninteracting Class-I model at V0=0.8 for q=7,9,11, tracking the first PT-breaking pair and the derivative dD_q/dγ at small γ up to the first crossing γ_c. If for any of these q the first PT-breaking pair is (1,2), or D_q is monotonic up to γ_pt, the general parity rule fails; if instead pair (2,3) breaks first and D_q is nonmonotonic, the rule is validated for q≥3 and the q=1 case should be stated as an explicit exception in the abstract and conclusion.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing gap is the generality of the odd-q branch of the parity rule. For q=3, Appendix B2 proves the inverted gap hierarchy Δ21>Δ32 and the (2,3) PT-breaking pair. For general odd q, the argument jumps at the sentence immediately before Eq. (B18): 'the lowest eigenstate is dominated by the self-conjugate center site in the lattice considered here.' This assertion is doing the real work: it guarantees the lowest band's first-order imaginary perturbation vanishes, that PT-breaking is delayed in the lowest band, and that the first exceptional point involves a higher pair. But it is not proved for q≥5, and it is not a generic property of self-conjugate sites. Appendix B4 shows that in phase Class III the parity rule reverses because the self-conjugate site is no longer in the lowest-energy sector. Within Class I itself, q=1 is an unqualified counterexample to the abstract's phrasing: Fig. 2(a) and the text report PT-breaking between the lowest two bands and monotonic D_1, not nonmonotonic E_2/E_3 behavior. The central mechanism for q=3,5 is numerically supported, but the claim 'odd denominators yield nonmonotonic response' is not established for general odd q and is literally false if q=1 is included.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper studies a one-dimensional PT-symmetric bichromatic optical lattice with commensurate ratio α=1/q and a weakly repulsive Bose-Einstein condensate. The authors compute noninteracting complex band structures by plane-wave diagonalization and report that the PT-breaking sequence and the response of the lowest moiré flat band to the imaginary potential depend on the parity of q: for even q the lowest two bands attract and break PT first, monotonically broadening the lowest band; for odd q (q=3,5 shown) the second and third bands break first while the lowest band remains real, producing a nonmonotonic dispersion D_q. A Wannier-overlap tight-binding model reproduces this picture and gives a parameter-free estimate of the q=2 threshold. The authors then solve the Gross-Pitaevskii equation self-consistently for weak repulsive interactions and find that interactions generally broaden the bands, with parity-dependent enhancement or suppression of flattening. A large-γ WKB analysis of effective hopping predicts D_q∼c_1√γ+c_2.","tokens_in":21167,"tokens_out":6943,"duration_ms":63197,"significance":"The paper is valuable because it connects a parity selection rule in a non-Hermitian commensurate superlattice to a concrete observable, the lowest-band width, and supports the rule with several independent methods: continuum diagonalization for q=1..6, perturbation theory, and a tight-binding model whose parameters are computed from Wannier overlaps rather than fitted. The tight-binding estimate γ_pt≈0.48 for q=2, close to the continuum value 0.46, is a genuine parameter-free check. The interacting GPE results for q=2..5 are a useful step toward experimental studies. The main risk is that the odd-q branch is only established for q=3 and q=5; the general statement in the abstract and conclusion needs qualification.","major_comments":[{"comment":"The parity statement is overgeneralized because q=1, which is an odd denominator under the definition α=1/q, q≥1, is treated in Fig. 2(a) as a single-lattice case in which PT-symmetry breaks between the lowest two bands and D_1 is monotonic in Fig. 2(g), not nonmonotonic with the lowest band remaining real. The abstract's claim that odd denominators 'yield a nonmonotonic response due to the PT-symmetry breaking within the second and the third lowest bands instead while the lowest band remains purely real' is therefore literally false if q=1 is included. The authors should either explicitly restrict the odd-denominator branch to q≥3 or redefine the parity rule for commensurate moiré supercells with q≥2, and adjust the conclusion in §V accordingly.","section":"Abstract and §III"},{"comment":"The general odd-q branch rests on an unproved assertion: 'the lowest eigenstate is dominated by the self-conjugate center site in the lattice considered here' immediately before Eq. (B18). This assertion is what guarantees that the first-order imaginary correction to the lowest band vanishes and that the first exceptional point involves the second and third bands. For q=3, Appendix B2 proves the inverted gap hierarchy Δ21>Δ32 via Eq. (B15), but for q≥5 the block-tridiagonal structure (B18) does not by itself imply center dominance, and no proof or numerical demonstration is given for q=5 or q=7. Since this is the load-bearing step for the general parity rule, I request either a tight-binding proof based on the real onsite-energy ordering and the smallness of the hoppings, or explicit numerical evidence of center dominance for q=5 and q=7 in the weak-γ regime.","section":"Appendix B3 (Eq. B18)"},{"comment":"The interacting odd-parity conclusion inherits the same gap: Figs. 4(c1) and 4(d1) show D_3 and D_5 only, and the statement in §V that 'for odd denominators, non-Hermiticity can either enhance or suppress the flattening' is presented as a general parity property. Because the underlying reason is the same center-dominance assumption, the interaction claim for q≥7 is not supported by the presented numerics. If the requested proof or evidence for odd q strengthens the noninteracting rule, the authors should explicitly state that the interacting extension is demonstrated for q=3 and q=5.","section":"§IV and Fig. 4"}],"minor_comments":[{"comment":"The label 'odd parities q_o={1,3,5}' is misleading because q=1 does not exhibit the odd-parity behavior described in the text; please relabel or add a caveat distinguishing q=1 as the single-lattice comparison limit.","section":"Fig. 2 caption"},{"comment":"The notation for the unperturbed wave function is incomplete: after writing 'where ˜ψ is the unperturbed wave function', the matrix element V_μν is not explicitly defined in terms of ψ̃_ν, and the normalization over the moiré cell should be stated.","section":"Eq. (13)"},{"comment":"The phrase 'G_q diverging at the EP' could be misread as if D_q diverges; since G_q=log10(w_q/Δ_q), it is the gap ratio that diverges as Δ_q→0. Please clarify.","section":"§III, after Eq. (12)"},{"comment":"The statement that the parity-dependent phenomenon 'can be understood by the perturbation theory' should explicitly mention that Eq. (13) is used for γ below the first exceptional point, since the expression is not valid near or beyond an EP.","section":"§III, perturbation-theory paragraph"},{"comment":"The main text does not mention that the potential in Eq. (6) corresponds to Class I of the relative-phase classification in Appendix B4; adding one sentence in §III would prevent readers from interpreting the parity rule as independent of the relative phase between primary and secondary lattices.","section":"Appendix B4"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is within the journal's scope and the numerical infrastructure is solid for the tested cases. The two issues to resolve are the q=1 overstatement in the abstract and the unproved center-dominance assumption that carries the general odd-q claim. Both are fixable with qualification or additional supporting evidence, so I recommend major revision rather than rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Read the abstract with a grain of salt, but the paper is worth your time. The genuinely new result is a parity-dependent PT-breaking sequence in 1D moiré lattices: even denominators couple the lowest two bands and break PT first (monotonic broadening), while odd denominators push the first breaking to the second/third-band pair and leave the lowest band real longer (nonmonotonic broadening). That is a structural insight I have not seen in the cited moiré or PT literature, and it is backed by continuum numerics for q=2–6, a parameter-free tight-binding model built from Wannier overlaps, and perturbation theory. The q=2 tight-binding threshold (0.48) matches the continuum (0.46) well, and the large-γ prediction of exp(−κ√γ) hopping is a clean testable statement.\n\nThe soft spots are real but not fatal. First, the abstract says odd denominators yield nonmonotonic response without exception, yet q=1 is odd and behaves like an even denominator — monotonic, PT breaking in the lowest two bands — as their own Fig. 2(a) shows. The text reports it correctly; the abstract overgeneralizes. Second, the general odd-q argument in Appendix B3 rests on the assertion that the lowest eigenstate is dominated by the self-conjugate center site. That is proven for q=3, but for q≥5 it is asserted, not derived, and it is doing load-bearing work. Numerics support q=5, so it may be true, but it is an assumption. Third, the parity rule is presented as a property of the denominator, but Appendix B4 shows it holds only for one of the four relative-phase classes; Class III reverses parity, and Classes II/IV erase the odd–even distinction. This scope is buried in an appendix. Minor: \"remains purely real\" is only valid up to the real-part crossing γ_c, after which band reordering occurs; that is explained in the text but not the abstract.\n\nNone of this breaks the core results for the model in Eq. (6). The paper is honest about its numerical method, the tight-binding parameters are computed rather than fitted, and the q=3 mechanism is explicitly derived. I would not cite the parity rule as a general theorem without the phase-class caveat, but within Class I it looks solid.\n\nThis is a job for a subfield referee: cold-atom moiré simulation and non-Hermitian band theory. It deserves a serious referee. Recommendation: send it out, but require (a) fixing the q=1 exception in the abstract, (b) proving or explicitly testing the center-dominance for general odd q, and (c) moving the phase-class limitation into the main text. With those changes the rule would be properly scoped.","headline":"New parity rule for PT-breaking in 1D moiré lattices, backed by numerics and parameter-free tight-binding, but the abstract overgeneralizes: q=1 is an exception, the odd-q mechanism is proven only for q=3, and the rule holds for only one relative-phase class.","tokens_in":21858,"tokens_out":6771,"would_cite":true,"duration_ms":54113,"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":"In a PT-symmetric one-dimensional moiré lattice for a weakly repulsive Bose-Einstein condensate, the parity of the ratio denominator $q$ determines whether an imaginary potential monotonically broadens the lowest flat band (even $q$) or…","keywords":["moiré lattices","PT symmetry","flat bands","bichromatic optical lattices","Bose-Einstein condensate","Gross-Pitaevskii equation","non-Hermitian band theory","commensurate ratios"],"falsifier":"Diagonalize the noninteracting continuum Hamiltonian at $V_0=0.8$ for $q=3$ with small $\\gamma$ and identify which bands first form a complex-conjugate pair: the parity rule predicts an exceptional point between bands 2 and 3 while band 1 stays real, so if the lowest two bands coalesce first, or if band 1 itself acquires an imaginary chemical potential at the same critical $\\gamma$, the central claim fails. A second check is to compute $D_q(\\gamma)$ for $q=5$ across $0<\\gamma<9$ and test the predicted nonmonotonic shape with a linear $\\sqrt{\\gamma}$ tail at large $\\gamma$.","tokens_in":20695,"feed_emoji":"🌀","tokens_out":16702,"duration_ms":130251,"temperature":0.7,"pith_summary":"This paper asks how the flatness of the lowest band in a one-dimensional moiré lattice is affected when the secondary lattice carries gain and loss, implemented as a $\\mathcal{PT}$-symmetric imaginary potential, and when the atoms interact weakly. The central claim is a parity rule: for commensurate ratios $\\alpha=1/q$ with even $q$, the lowest two bands are directly coupled by the imaginary potential, $\\mathcal{PT}$ symmetry breaks between them, and the lowest band broadens monotonically as the gain-loss strength $\\gamma$ grows; for odd $q$, a central lattice site at which the imaginary potential vanishes shields the lowest band, so $\\mathcal{PT}$ symmetry first breaks between the second and third bands while the lowest band remains real, and its width responds nonmonotonically, first increasing and then decreasing. The paper derives this rule from second-order perturbation theory plus a tight-binding analysis, and verifies it numerically for $q=1$ through $q=6$. It then solves the Gross-Pitaevskii equation for weak repulsive interactions and claims that interactions alone broaden the bands, yet the parity classification survives: even $q$ uniformly loses flattening as $\\gamma$ grows, whereas odd $q$ shows an interplay in which the imaginary potential can either enhance or reduce flattening. If correct, the result gives a simple design rule for controlling moiré band flatness with engineered dissipation in ultracold-atom simulators.","feed_headline":"Even moiré ratios broaden a flat band under loss; odd ones protect it","feed_subtitle":"The parity of the ratio's denominator sets which bands break PT symmetry first and whether the lowest band stays flat.","key_machinery":"The load-bearing object is the $\\mathcal{PT}$-conjugate pairing of the $q$ Wannier centers inside one moiré cell and the presence or absence of a self-conjugate central site. For even $q$ the centers split into $q/2$ conjugate pairs, and the lowest two bands form a bonding-antibonding doublet whose states are concentrated on one pair; the paper's perturbative identity $E_\\nu=\\tilde{E}_\\nu+\\gamma^2\\sum_{\\mu\\ne\\nu}|V_{\\mu\\nu}|^2/\\Delta_{\\mu\\nu}+O(\\gamma^3)$, with $V_{\\mu\\nu}=-V_{\\nu\\mu}^*$, then shows that the smallest gap and largest imaginary-potential overlap make the lowest two bands attract most strongly, producing the monotonic broadening. For odd $q$ a single center site lies at the cell center with vanishing imaginary potential; because the lowest eigenstate is concentrated there, the first $\\mathcal{PT}$-breaking transition is moved to the second and third bands, and the lowest band's leading imaginary correction is second order and real. At large $\\gamma$ the paper uses an effective-hopping argument with left and right non-Hermitian Wannier functions to show that the lowest bandwidth decays as $\\exp(-\\kappa\\sqrt{\\gamma})$, with $\\kappa=q\\sqrt{V_s/8}\\int_0^{2\\pi}\\sqrt{|\\sin u|}\\,du$, giving $D_q\\propto-\\sqrt{\\gamma}$; this is what forces the odd-$q$ response to turn around and flatten again.","core_discovery":"The paper's central claim is that the parity of the denominator $q$ in the commensurate moiré ratio $\\alpha=1/q$ determines the $\\mathcal{PT}$-breaking sequence and thereby the fate of the lowest flat band under a $\\mathcal{PT}$-symmetric imaginary potential. For even $q$, every potential minimum within the moiré cell belongs to a $\\mathcal{PT}$-conjugate pair, and the lowest two bands form a directly coupled doublet; the imaginary potential induces level attraction between them, so they coalesce at an exceptional point and the lowest band's width increases monotonically with $\\gamma$. For odd $q$, one minimum lies at the cell center where the imaginary potential is exactly zero, and the lowest eigenstate is dominated by this self-conjugate site; its first-order coupling to gain and loss vanishes, so $\\mathcal{PT}$ symmetry first breaks between the second and third bands, the lowest band remains purely real over a broad range of $\\gamma$, and its width first grows and then shrinks again because level attraction from higher bands competes with non-Hermitian confinement that exponentially suppresses the effective hopping at large $\\gamma$. For a weakly repulsive condensate solved through the Gross-Pitaevskii equation, the paper finds that interactions broaden the bands by themselves but preserve the parity dependence: for even $q$ the lowest band consistently becomes less flat with $\\gamma$, whereas for odd $q$ the imaginary potential can either enhance or suppress flattening depending on the parameter regime. The same tight-binding analysis shows that the parity rule can reverse if the relative lattice phases place the self-conjugate site outside the lowest-energy sector.","pith_inferences":["Beyond the paper, the even-odd distinction gives a clear experimental fingerprint: the lowest-band width versus gain-loss strength should grow monotonically for $q=2$ but show a dip-and-revival for $q=3$, so one transport or band-mapping measurement could test the mechanism directly.","Beyond the paper, the large-$\\gamma$ result $D_q\\simeq c_1\\sqrt{\\gamma}+c_2$ with $c_1<0$, demonstrated for $q=5,7$, suggests a universal stretched-exponential localization of non-Hermitian Wannier tails for all odd $q$, testable by exact diagonalization for larger $q$.","Beyond the paper, the phase-class analysis implies that the relative phase between the two lattices can swap even and odd behavior, turning the parity rule into a two-state switch for protecting or broadening the lowest band.","Beyond the paper, the predicted band asymmetry $E_k\\ne E_{-k}$ for interacting condensates implies nonreciprocal transport, a consequence the paper does not develop; expansion or Bloch-oscillation experiments could look for direction-dependent group velocities."],"forward_implications":["For even $q$, increasing the gain-loss strength $\\gamma$ monotonically increases the lowest-band width $D_q$ and the inverse gap ratio $G_q$ up to the exceptional point, so dissipation acts as a linear knob for band broadening.","For odd $q$, the lowest band stays real even after $\\mathcal{PT}$ symmetry breaks in higher bands, and $D_q$ is nonmonotonic in $\\gamma$; at large $\\gamma$ the band re-flattens with $D_q\\simeq c_1\\sqrt{\\gamma}+c_2$, $c_1<0$.","Weak repulsive interactions broaden the lowest band on their own, but in the combined system the even-$q$ case always becomes less flat with $\\gamma$, while the odd-$q$ case shows either enhancement or suppression of flattening.","With interactions, the mean-field band becomes asymmetric ($E_k\\ne E_{-k}$) and the ground state shifts away from $k=0$ while the band maximum shifts away from the zone boundary; both effects grow with interaction strength and $\\gamma$.","The tight-binding analysis implies the parity rule is tied to the self-conjugate site sitting in the lowest-energy sector, so changing the relative primary-secondary lattice phase (Class I versus Class III) swaps the even-odd behavior."],"supporting_citations":[{"why":"Supplies the 1D moiré lattice model and the exponentially flattened lowest band that defines the quantity whose fate is studied.","marker":"[10]"},{"why":"Provides the nonlinear PT-symmetric lattice reference with swallowtail bands that motivates combining interactions with PT symmetry.","marker":"[42]"},{"why":"Gives the coordinate transformation fixing the PT threshold gamma_pt = 2 for the single-lattice case q = 1.","marker":"[36]"},{"why":"Supplies the non-Hermitian tight-binding formalism, PT-pairing relations, and the Peierls transformation used in Appendix B.","marker":"[26]"},{"why":"Establishes the PT-symmetric double-well pairing picture used to explain the even-q attraction between the lowest two bands.","marker":"[46]"},{"why":"Gives the 1D reduction of the Gross-Pitaevskii equation used for the interacting condensate.","marker":"[43]"},{"why":"Defines the maximally localized Wannier functions used to construct the tight-binding moiré model in Appendix B.","marker":"[72]"},{"why":"Provides the non-Hermitian Wannier and effective-hopping framework used for the large-gamma flattening of odd-q bands.","marker":"[80]"}],"fun_headline_variants":["Moiré flat band: odd ratios resist loss, even ones yield","Parity of moiré ratio governs flat band under PT loss","Even moiré ratios flatten less under loss; odd can stay flat","PT loss broadens even-parity flat bands, spares odd ones","Weak loss: odd moiré ratios keep flat band, even lose it"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"For odd $q$, the entire parity rule rests on the assumption that the lowest eigenstate is dominated by the self-conjugate central lattice site, where the imaginary part of the potential vanishes; if interactions, deeper lattices, or different lattice phases move the lowest state onto a conjugate pair, the predicted protection of the lowest band is lost and the even-odd behavior can reverse.","fun_headline_variants_meta":{"raw":{"variants":["Moiré flat band: odd ratios resist loss, even ones yield","Parity of moiré ratio governs flat band under PT loss","Even moiré ratios flatten less under loss; odd can stay flat","PT loss broadens even-parity flat bands, spares odd ones","Weak loss: odd moiré ratios keep flat band, even lose it"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000303,"raw_usage":{"total_tokens":1888,"prompt_tokens":1235,"completion_tokens":653,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":851,"completion_tokens_details":{"reasoning_tokens":557}},"tokens_in":851,"tokens_out":653,"duration_ms":7037,"temperature":1.0,"reasoning_tokens":557,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T00:09:46.169516+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Diagonalize the noninteracting continuum Hamiltonian at $V_0=0.8$ for $q=3$ with small $\\gamma$ and identify which bands first form a complex-conjugate pair: the parity rule predicts an exceptional point between bands 2 and 3 while band 1 stays real, so if the lowest two bands coalesce first, or if band 1 itself acquires an imaginary chemical potential at the same critical $\\gamma$, the central claim fails. A second check is to compute $D_q(\\gamma)$ for $q=5$ across $0<\\gamma<9$ and test the predicted nonmonotonic shape with a linear $\\sqrt{\\gamma}$ tail at large $\\gamma$.","supporting_citations":[{"cited_title":"Midya, B","cited_arxiv_id":null,"evidence_quote":"Provides the nonlinear PT-symmetric lattice reference with swallowtail bands that motivates combining interactions with PT symmetry."},{"cited_title":"Graefe and H","cited_arxiv_id":null,"evidence_quote":"Gives the 1D reduction of the Gross-Pitaevskii equation used for the interacting condensate."}],"review_version":2}