{"id":"e290250b-e23d-498b-9630-ff7f53e0be03","arxiv_id":"1908.02774","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Generic quasiperiodic spin chains flow under real-space renormalization to discrete Fibonacci-like sequences, making their quantum critical behavior exactly tractable.","lead":"This paper shows that many quasiperiodic spin chains, whose couplings vary smoothly and never quite repeat, behave under renormalization as if their couplings followed simple Fibonacci-like repeating patterns. This makes their quantum critical points solvable and yields exact exponents for Heisenberg, Ising, Potts, and a model of many-body localization.","discovery_kind":"unification","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The Fibonacci RG step decimates all minima at once, but adjacent B bonds in ABABA share an A bond that the first decimation removes, so Eq. (4) is not derived from the standard sequential SDRG and the flow to Fibonacci sequences is not yet established.","rationale":"The paper's strongest claim is that generic quasiperiodic modulations flow under real-space RG to discrete Fibonacci-like substitution sequences, enabling asymptotically exact critical exponents. The load-bearing pillar of this claim is the simultaneous decimation of all local-minimum bonds in a Fibonacci step. The reader's weakest assumption already identified this step as uncontrolled when neighboring coupling ratios are not large. My analysis sharpens that concern into a concrete structural objection: in the ABABA pattern, two B bonds share an A bond, and the standard SDRG consumes that shared A bond in the first decimation. The simultaneous update therefore double-counts it and is not the strong-disorder RG whose asymptotic exactness is claimed. The proof of sharpening and the asymptotic formula A_m-B_m=mc are derived inside the simultaneous scheme, so they do not address the overlap error. The numerical checks in Fig. 1 use the same simultaneous step, so they cannot validate equivalence to sequential SDRG. The proposed test directly compares the two RG procedures. If the test passes, the flow and exponents survive; if it fails, the central universality claim needs a different, controlled justification. The reader's CONDITIONAL verdict remains appropriate because the concern is serious but testable, and the paper has independent numerical support for specific exponents that a fix might preserve. I therefore recommend no change to the reader's verdict.","tokens_in":19145,"tokens_out":10543,"duration_ms":124217,"concrete_test":"Run the standard one-bond-at-a-time SDRG on a Heisenberg chain of length L=F_{3l} (for example, 34, 144, or 610) with l_i=1+cos(2*pi*phi*i) for golden-ratio phi. At each step decimate the global minimum of l, update the two neighboring bonds using Eq. (2) with c=ln 2, and continue until the chain is fully decimated. After each set of decimations that reduces the chain length by a factor of phi^3, compare the remaining l-values with the A_m and B_m bands predicted by Eq. (4), and check whether the decimated bonds in each block coincide with the B positions of the Fibonacci word. If the decimation order deviates from the Fibonacci pattern or the l-values differ from Eq. (4) by more than a few percent at any stage, the simultaneous Fibonacci step is not equivalent to the standard SDRG, and the central universality claim loses its asymptotic-exactness support.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central reduction rests on the claim that one can decimate every local-minimum bond B_m in a single Fibonacci step, producing the renormalized couplings A_{m+1}(n)=A_m(n-2)-B_m(n-1)+A_m(n)-B_m(n+1)+A_m(n+2)+2c and B_{m+1}(n)=A_m(n-1)-B_m(n)+A_m(n+1)+c (Supplement I.B). In the Fibonacci word the pattern ABABA occurs, where the two B bonds at n-1 and n+1 share the middle A bond at n. The standard strong-disorder RG decimates one bond at a time: decimating B(n-1) removes bonds n-2, n-1, and n, and creates a new bond between spins n-2 and n+1. The shared A(n) is consumed by that first decimation, so the later decimation of B(n+1) acts on a different neighboring bond and produces a different effective coupling. The simultaneous update double-counts the shared A bond and is not equivalent to the SDRG whose asymptotic exactness is being invoked. The sharpening of the bands and the asymptotic formula A_m-B_m=mc+o(1) are derived inside this approximate simultaneous scheme; the error from overlapping decimations is never controlled. Consequently, Eq. (4), the fixed-point Fibonacci sequence, and the exponents nu=1 and Delta_sigma are not yet shown to follow for Hamiltonian (1). This concern applies already to the cosine potential, independently of the broader basin-of-attraction claim for generic quasiperiodic modulations. The numerical demonstration in Fig. 1 implements the same simultaneous step, so it cannot resolve the equivalence issue.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies one-dimensional quantum spin chains with quasiperiodically modulated couplings. Its central claim is that, under the standard real-space strong-disorder renormalization group, generic smooth quasiperiodic potentials with golden-ratio frequency flow to discrete self-similar Fibonacci-like substitution sequences, which act as fixed points of a functional renormalization group. This is used to obtain asymptotically exact critical properties of the Heisenberg, Potts, and Ising chains and of a phenomenological quasiperiodic MBL transition model, including the correlation-length exponent ν=1 and a Potts spin exponent Δσ. The Supplement derives exact band formulas for a cosine potential, proves the Fibonacci pattern of local minima for that potential, and provides numerical evidence for exponential sharpening of the bands.","tokens_in":19530,"tokens_out":18010,"duration_ms":198880,"significance":"If fully established, the result would substantially generalize earlier work on binary substitution sequences to a broad class of smooth quasiperiodic modulations and would provide a rare example of an asymptotically exact treatment of quasiperiodic quantum criticality. The cosine-potential analysis is concrete, the claimed exponents are falsifiable, and the paper is candid about the uncontrolled Ising case and the toy-model status of the MBL analysis. The main weakness is that the basin of attraction of the Fibonacci sequence is not proven for the full class of potentials for which universality is claimed.","major_comments":[{"comment":"The statement that the flow to Fibonacci-like sequences holds for 'any sufficiently regular quasiperiodic potential with frequency φ' is not supported by the proofs in the Supplement and is in tension with the Supplement's own observation that f(x)=sin(4x) destroys the Fibonacci structure. The analytic argument covers the cosine potential and, via a monotonicity argument, potentials of the form f(cos(2πφn+θ)) with f monotonic on (-1,1). Because the basin of attraction is load-bearing for the paper's universality claim, the authors should either prove the flow for a precisely defined larger class or explicitly restrict the claim to the class for which analytic or systematic numerical evidence exists.","section":"Main text 'Flow to discrete sequences'; Supplement I.B 'Generic potentials'"},{"comment":"The claim that the site-dependent fluctuations ϵ_{i,m} decay exponentially as m→∞ is used to conclude that the decimation rule becomes asymptotically exact, but this decay is demonstrated only numerically (Fig. 1b) and via a heuristic 'longer patterns give tighter phase constraints' argument. An analytic bound on the bandwidths of A_m and B_m, or at least a precise statement that the exponential decay is a numerical finding, is needed to support the 'asymptotically exact' characterization.","section":"Supplement I.B and main text Eq. (4)"},{"comment":"The spin exponent Δσ in Eq. (6) is derived analytically for the Fibonacci fixed point in which h_i and J_i are described by the same Fibonacci sequence, but the main text applies it to the more natural case of distinct quasiperiodic potentials with a relative phase. The Supplement notes that in that case the fixed point oscillates among multiple sequences and the equality of exponents is established only numerically (Fig. 2). The main text should state this qualification when presenting Eq. (6), rather than presenting the formula as applying directly to the distinct-potential case.","section":"Main text 'Quantum Potts model'; Supplement I.D"}],"minor_comments":[{"comment":"The simultaneous decimation of all B bonds is not obviously equivalent to the standard sequential SDRG when two B bonds are separated by a single A bond. I checked the explicit example ABABA: sequentially decimating B_{n-1} and then B_{n+1} yields exactly the formula for A_{m+1}(n) given in the Supplement, so the feared double-counting of the intervening A bond does not occur. The authors should add a sentence stating this equivalence, since the cluster decimation step is central to the derivation.","section":"Supplement I.B"},{"comment":"The expression for Δσ is ambiguous as typeset: it should be made clear whether the argument of the logarithm is 1 + 2φ^{-1}/3 or 1 + 2φ^{-1/3}; the two readings give different numerical exponents.","section":"Main text Eq. (6)"},{"comment":"There are several typographical errors: 'an Heisenberg chain' in the Fig. 1 caption, 'pannel' in the Supplement, and 'satifies' in the main text near the MBL defect argument.","section":"Throughout"},{"comment":"The caption text says 'The red dot is a local minimum insulating block while the green dot corresponds to the adjacent thermal block,' but the main text example only discusses the red dashed line; the notation in the figure should be aligned with the definitions in the text.","section":"Supplement II.A.2, Fig. 5"},{"comment":"The phrase 'nonsingular sequences of 𝓁 are generically squashed under the RG' would benefit from a precise definition of 'squashed', since this is the mechanism behind the marginal irrelevance claim.","section":"Main text 'Ising model'"}],"recommendation":"major_revision","confidential_remarks":"The core cosine-potential derivation is strong and the paper is likely to be publishable after the universality claims are tightened to match the proofs. The main risk is the overbroad basin-of-attraction claim; the authors should either prove it for a defined class or reframe the paper as establishing the flow for cosine-type potentials with numerical evidence for a wider class."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe one thing you should know: the central claim—generic smooth quasiperiodic modulations flow under strong-disorder RG to Fibonacci-like discrete sequences—is genuinely new and, for the cosine potential, backed by a real proof in the supplement. I was initially worried by the stress-test note about simultaneous decimation double-counting shared A bonds in ABABA clusters. It doesn't hold up. If you decimate B_{n-1} sequentially, the new bond reads A_{n-2}-B_{n-1}+A_n+c; using that bond as the left neighbor when decimating B_{n+1} gives exactly their A_{m+1} update. The shared A appears exactly once. The intermediate bond is much weaker than the B's, so the simultaneous step is the same as two sequential steps to leading order.\n\nWhat the paper does well: the proof that cosine local minima form the Fibonacci word is clean; the band-sharpening calculation and the exponential decay of fluctuations are concrete; the ν=1 argument from the dimerization eigenvalue λ=φ^3 is simple and convincing. Citations are appropriate—they credit earlier Fibonacci-chain and free-fermion work, and self-citations are to their disclosed MBL RG model.\n\nThe soft spots are real but fixable. First, the Potts spin exponent in Eq. (6) is inconsistent with the supplement's own derivation. From their P(r) formula, Δσ = -ln(3φ/(3φ+2))/(3 lnφ) ≈ 0.238, which matches their numerical L^{-0.47}. Eq. (6)'s ln(1+2φ^{-1/3})/(3lnφ) gives ≈ 0.688. This looks like a typo, but it's in both the main text and the supplement, so a referee should flag it. Second, the universality claim for 'any sufficiently regular quasiperiodic potential' is not proven; the exact argument covers only the cosine potential. The supplement's monotonic-function extension is plausible and numerics help, but the basin of attraction is asserted more broadly than demonstrated. Third, the simultaneous decimation is controlled only once bands are separated; early steps rely on the numerical observation of small fluctuations. For a letter, that's acceptable.\n\nWho is this for? People working on quasiperiodic quantum criticality, and to a lesser extent MBL phenomenology. I'd send it to a serious referee. The main mechanism is credible, the errors are corrigible, and the idea will be useful even if the strong universality claim gets tightened.","headline":"The flow-to-Fibonacci idea is new and survives the sequential-decimation objection; the real issues are the Potts exponent typo and an overbroad universality claim.","tokens_in":19980,"tokens_out":9017,"would_cite":true,"duration_ms":83737,"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":"For a wide class of spin chains, generic quasiperiodic modulations flow under real-space renormalization to discrete Fibonacci-like substitution sequences, making quasiperiodic quantum critical points asymptotically exactly solvable.","keywords":["quasiperiodic spin chains","real-space renormalization group","Fibonacci sequence","substitution sequences","quantum criticality","Harris-Luck criterion","many-body localization transition","correlation length exponent"],"falsifier":"Directly iterate the decimation rule on the cosine potential and measure the spread of couplings within the $A$ and $B$ bands: if the spread does not decay exponentially with the number of Fibonacci steps, or if the energy gap of a chain of length $L$ does not scale as $\\Delta E \\sim e^{-c(3\\ln\\varphi)^{-2}\\ln^2 L}$, the central claim fails.","tokens_in":18965,"feed_emoji":"🧲","tokens_out":11118,"duration_ms":102047,"temperature":0.7,"pith_summary":"The paper argues that generic quasiperiodic modulation of couplings, meaning smooth functions with golden-ratio periodicity, is not an intractable complication. Under a real-space decimation renormalization group, such potentials flow to discrete Fibonacci-like substitution sequences, which act as fixed points of a functional renormalization group. This flow makes quasiperiodic quantum critical points asymptotically exactly solvable: the paper derives the correlation length exponent $\\nu=1$ for the Heisenberg and Potts chains and for a quasiperiodic many-body localization transition model, an infinite dynamical exponent for the Heisenberg chain, and an analytic spin-scaling dimension for the Potts chain. If the central claim is right, previously isolated results on discrete aperiodic sequences apply to every sufficiently regular quasiperiodic potential.","feed_headline":"Generic quasiperiodic spin chains flow to Fibonacci fixed points","feed_subtitle":"It makes quasiperiodic critical points exactly solvable, yielding exponent ν=1 for Heisenberg, Potts, and MBL.","key_machinery":"The central object is the Fibonacci RG step. In the log-coupling variables $\\ell_j = -\\ln J_j$, the real-space decimation rule is $\\ell'_i = \\ell_{i-1} - \\ell_i + \\ell_{i+1} + c$, where $c=\\ln 2$ for the Heisenberg chain and $c=\\ln(q/2)$ for the Potts chain. For any smooth golden-ratio potential, the local minima of $\\ell$ form the Fibonacci word; labeling minima as $B$ and the rest as $A$, decimating all $B$ couplings at once maps the system to a new Fibonacci word via the inflation rule $A\\to AB$, $B\\to A$, or equivalently $ABABA\\to A$ and $ABA\\to B$. Iterating this step sharpens the $A$ and $B$ bands and drives fluctuations to zero exponentially, turning the heuristic decimation into an asymptotically exact description.","core_discovery":"The central claim is that, for a wide class of one-dimensional spin chains, every sufficiently regular quasiperiodic modulation flows under real-space renormalization to a perfect binary Fibonacci sequence or a related substitution sequence, so that discrete-sequence fixed points control generic quasiperiodic criticality rather than only specially tailored couplings. The discovery is the flow itself: starting from $\\ell_j = a + \\cos(2\\pi\\varphi j + \\theta)$, after $m$ simultaneous decimation steps the couplings collapse into two bands $A_m$ and $B_m$ whose intra-band fluctuations vanish exponentially, with $A_m - B_m \\approx m c$, and the decimation rule becomes asymptotically exact. The fixed-point sequence then determines the critical properties: the Heisenberg chain has dynamical exponent $z=\\infty$ and correlation length exponent $\\nu=1$; the $q$-state Potts chain with $q>2$ also has $\\nu=1$ plus an analytic order-parameter scaling dimension; and the quasiperiodic many-body localization toy model has $\\nu=1$ from a defect argument based on Fibonacci approximants.","pith_inferences":["Our inference: the same functional-RG logic should apply to other real-space decimation rules with a positive additive constant, such as generalized random-singlet rules in other geometries, predicting flow to substitution sequences whose inflation rules are set by the local-minimum pattern rather than by the microscopic Hamiltonian.","Our inference: the defect argument for $\\nu=1$ suggests a way to engineer other exponents deliberately; by inserting square-root singularities at the Fibonacci-minimum boundaries, the paper's own calculation gives $\\nu=2$, so a family of singular functions could interpolate continuously between exponents.","Our inference: since quasiperiodic systems lack rare thermal regions, the toy-model result supports the scenario of two distinct universality classes for the many-body localization transition, and an exact-diagonalization study of a quasiperiodic chain measuring the correlation-length exponent could distinguish $\\nu=1$ from the random-case value.","Our inference: the flow-to-sequence mechanism is probably not confined to one dimension; applying the same simultaneous-decimation idea to two-dimensional quasiperiodic tilings, where local minima form more complex substitution rules, would be a natural test of whether the universality survives."],"forward_implications":["Smooth quasiperiodic modulation of Heisenberg and $q>2$ Potts chains becomes exactly tractable: all such potentials with golden-ratio frequency share the same critical exponents, $\\nu=1$ and $z=\\infty$ for Heisenberg, and the Potts spin-scaling dimension $\\Delta_\\sigma = \\ln(1+2\\varphi^{-1/3})/(3\\ln\\varphi)$.","The quasiperiodic many-body localization transition in the toy RG has correlation length exponent $\\nu=1$ for smooth initial conditions, distinct from the random case, and the dimerization perturbation is dangerously irrelevant at the critical sequence.","For the Ising chain, where the decimation constant vanishes, smooth quasiperiodic modulation is marginally irrelevant and is squashed to a constant, so nontrivial quasiperiodic criticality requires a singular distribution of couplings and yields a finite dynamical exponent $z\\approx 1.6$ in the RG.","Potts chains with different relative phases between the field and coupling sequences can be attracted to different self-similar sequences, but all observed fixed points share $\\nu=1$ and the same spin-scaling dimension.","For irrational frequencies beyond the golden ratio, the fixed point is not self-similar under a single RG step; each level is governed by the continued-fraction expansion, so the sequence fixed point becomes level-dependent."],"supporting_citations":[{"why":"Supplies the Harris-Luck criterion $\\nu\\ge 1$ that determines when quasiperiodic modulation is relevant at a critical point.","marker":"[18]"},{"why":"Establishes the infinite-randomness fixed point and the strong-disorder decimation scheme that the quasiperiodic analysis adapts.","marker":"[2]"},{"why":"Gives the real-space decimation rule for random spin chains that is iterated in the quasiperiodic case.","marker":"[34, 35]"},{"why":"Provides earlier exact Fibonacci-sequence results for the Heisenberg chain that the paper extends to generic potentials.","marker":"[25]"},{"why":"Supplies the earlier treatment of generic quasiperiodic potentials for free fermions and the continued-fraction structure for general irrational frequencies.","marker":"[27]"},{"why":"Defines the thermal-insulating block RG whose quasiperiodic version is analyzed for the MBL transition.","marker":"[31]"},{"why":"Introduces the asymmetry parameter $\\beta$ and the Kosterlitz-Thouless-type random MBL transition that the quasiperiodic analysis modifies.","marker":"[32]"},{"why":"Contains the proof that local minima of a cosine potential form the Fibonacci word and the derivations of the exponents quoted in the main text.","marker":"[38]"}],"fun_headline_variants":["Quasiperiodic criticality pinned down exactly","Generic quasiperiodic chains become exactly solvable","Fibonacci fixed points explain quasiperiodic criticality","Universal ν=1 across quasiperiodic spin chain classes"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument depends on the assumption that decimating all local-minimum bonds at once stays accurate even before the coupling ratios become large, and that every sufficiently regular quasiperiodic pattern flows to the Fibonacci sequence even though the detailed proof is given only for a cosine potential.","fun_headline_variants_meta":{"raw":{"variants":["Quasiperiodic criticality pinned down exactly","Generic quasiperiodic chains become exactly solvable","Fibonacci fixed points explain quasiperiodic criticality","Universal ν=1 across quasiperiodic spin chain classes"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001029,"raw_usage":{"total_tokens":4318,"prompt_tokens":911,"completion_tokens":3407,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":527,"completion_tokens_details":{"reasoning_tokens":3346}},"tokens_in":527,"tokens_out":3407,"duration_ms":25539,"temperature":1.0,"reasoning_tokens":3346,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T14:35:27.730326+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Directly iterate the decimation rule on the cosine potential and measure the spread of couplings within the $A$ and $B$ bands: if the spread does not decay exponentially with the number of Fibonacci steps, or if the energy gap of a chain of length $L$ does not scale as $\\Delta E \\sim e^{-c(3\\ln\\varphi)^{-2}\\ln^2 L}$, the central claim fails.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the Harris-Luck criterion $\\nu\\ge 1$ that determines when quasiperiodic modulation is relevant at a critical point."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes the infinite-randomness fixed point and the strong-disorder decimation scheme that the quasiperiodic analysis adapts."},{"cited_title":"Hida, Physical Review Letters 93, 037205 (2004)","cited_arxiv_id":null,"evidence_quote":"Provides earlier exact Fibonacci-sequence results for the Heisenberg chain that the paper extends to generic potentials."},{"cited_title":"Wilkinson, Proceedings of the Royal Society of Lon- don","cited_arxiv_id":null,"evidence_quote":"Supplies the earlier treatment of generic quasiperiodic potentials for free fermions and the continued-fraction structure for general irrational frequencies."},{"cited_title":"Zhang, B","cited_arxiv_id":null,"evidence_quote":"Defines the thermal-insulating block RG whose quasiperiodic version is analyzed for the MBL transition."},{"cited_title":"Goremykina, R","cited_arxiv_id":null,"evidence_quote":"Introduces the asymmetry parameter $\\beta$ and the Kosterlitz-Thouless-type random MBL transition that the quasiperiodic analysis modifies."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Contains the proof that local minima of a cosine potential form the Fibonacci word and the derivations of the exponents quoted in the main text."}],"review_version":1}