{"id":"809969aa-3b94-47ee-81ce-7d0938be10b3","arxiv_id":"2607.26158","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"In the slow-fast limit, the phase-lock areas of the RSJ Josephson junction model converge to a parquet of integer-spaced squares/strips with boundaries parallel to u±ℓ=0.","lead":"For the overdamped Josephson junction model, this paper proves that as the driving frequency becomes very small and the parameters approach a critical point, the regions of frequency locking converge to a simple tiling of the parameter plane by squares, strips, and a sector. The result answers an open question from 2010 and extends to a whole class of slow-fast systems on a torus.","discovery_kind":"first_principles","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Lower inclusion rests on unproved stable-flowbox lemma (Prop. 2.19), cited to [28] for a subfamily; if the cited result does not yield the required uniform exponential contraction for all compact parameter rectangles, the parquet proof has a gap.","rationale":"After reading the proof in detail, I find no internal contradiction in the main argument. The Riccati analysis (Propositions 2.13–2.15), the sector-crossing estimates (Proposition 2.23, Claims 2.24–2.26), and the upper-inclusion argument in Theorem 1.11 are coherent and rely on standard comparison and compactness arguments. The convergence to the stable solution x_-(y) via Lemma 2.11 is well supported. However, the entry into the first Riccati chart and the subsequent drifts along the slow curve are justified only by Proposition 2.19, which is an unproved external citation. The paper itself (Remark 1.15) suggests the cited work covers only a subfamily of the RSJ family, so the generality of Proposition 2.19 is in doubt. This is exactly the reader's weakest assumption. The typos (Theorem 1.11's ℓ,u formulas, the symmetry sign in the proof of Proposition 1.18, and the '(1,1)' vertex) do not affect the RSJ theorem. Because the missing lemma is likely true and standard, the correct verdict remains CONDITIONAL: the authors should either supply a proof of Proposition 2.19 for the stated generality or restrict the citation and show it applies uniformly on the relevant parameter rectangles. I therefore do not change the reader's verdict.","tokens_in":36946,"tokens_out":18170,"duration_ms":154696,"concrete_test":"Check whether [28, Thm 3, Prop 4] implies Proposition 2.19 for all systems (1.13) satisfying conditions (i)–(vi) of Theorem 1.11, with constants d>0 uniform on the compact parameter sets Z_{m,ε}×Z_{k,ε}. Since Remark 1.15 indicates [28] treats only a subfamily of (1.17), the most direct test is to prove Proposition 2.19 from scratch for the RSJ family (1.17): linearize around the straight slow curve C_{ℓ,u;ω}, compute the contraction exponent ∫ (∂f/∂θ) dτ along the slow curve, and verify the flowbox width is O(exp(−d/ω)) with d>0 independent of (ℓ,u) in the preimage of Z_{m,ε}×Z_{k,ε}. If this computation succeeds and yields the uniform bound, the gap is benign and the proof can be completed; if it reveals a parameter region where the exponent vanishes, the lower inclusion fails.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The proof of the lower inclusion of Theorem 1.2 and of the general Theorem 1.11 hinges on Proposition 2.19, which asserts that a segment transverse to the slow curve evolves into a stable flowbox O(ω)-close to the slow curve with width ≤ exp(−d/ω), uniformly for parameters in compact subsets. This proposition is not proved; it is dismissed with 'The proposition follows from the classical theory of slow-fast systems, see, e.g., [28, theorem 3 and proposition 4]' (end of §2.5). It is used at the entry to the first Riccati neighborhood (Lemma 2.16, Step 3), again in the drift along the stable graph to the target segment (Step 5 and Claim 2.27, where it is called 'well-known'), and in Remark 1.15. If the exponential contraction or the O(ω)-closeness fails, or if the cited result does not actually cover the class (1.13) under conditions (i)–(vi), the square-filling argument collapses. The concern is reinforced by Remark 1.15, which states that [28] studies 'a subfamily of (1.17)' for canards—i.e., not necessarily the full generality of Theorem 1.11. For the RSJ case the stable graph has no horizontal segments, so Claim 2.27 reduces to Prop. 2.19, but even for RSJ the uniform estimate over Z_{m,ε}×Z_{k,ε} is not derived. This is an external-support gap, not an internal inconsistency; it is addressable by a proof or a precise reference.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the overdamped Josephson junction (RSJ) model in the slow-fast limit ω→0 while (B,A)=(ℓω,1+uω). In the rescaled parameters (ℓ,u), it claims that the phase-lock area L_r(ω) converges to an explicit piecewise-linear parquet L^0_r: an infinite chain of unit squares whose vertices are integer lattice points on the line ℓ=r, together with a strip or sector. It further claims that the vertices of L^0_r are exactly the limits of constrictions of L_r(ω). The proof proceeds by rescaling near the Morse critical points of the slow curve to a Riccati equation, analyzing its stable and unstable solutions via the quantum harmonic oscillator, and then assembling a periodic orbit through a sequence of stable flowboxes. A generalization, Theorem 1.11, is stated for a class of slow-fast systems on T^2 with two Morse critical points and a stable slow-fast graph.","tokens_in":37339,"tokens_out":4731,"duration_ms":42412,"significance":"If the proof is completed, this is a substantial result: it gives the first rigorous asymptotic portrait of Arnold tongues for the RSJ model in the slow-fast limit, resolving a long-standing open question of Buchstaber. The limiting parquet is derived rather than fitted, and it yields falsifiable, parameter-free predictions about phase-lock areas and constrictions. The paper also contains a nontrivial generalization to a class of slow-fast systems, and the Riccati pole-counting argument is elegant and well documented. The main risk is not the plausibility of the conclusion but the dependence of the lower-inclusion proof on an unproved, externally cited stable-flowbox lemma.","major_comments":[{"comment":"The lower inclusion of Theorem 1.2 and the general Theorem 1.11 rests on Proposition 2.19, which asserts that a transverse segment evolves into a flowbox O(ω)-close to the slow curve with width ≤ exp(−d/ω), uniformly over compact parameter sets. This proposition is not proved; the text refers to [28, Theorem 3 and Proposition 4] and to 'classical theory'. However, Remark 1.15 explicitly notes that [28] studies only a subfamily of (1.17), and the uniform estimate over Z_{m,ε}×Z_{k,ε} is not derived. The same issue appears in Claim 2.27, whose proof is one sentence: 'well-known from slow-fast theory' and which invokes Proposition 2.19 only when the stable graph has no horizontal segments, whereas the general graph may contain them. Since Lemma 2.16, Steps 3–5, and Claim 2.27 all use this estimate to carry orbits from one critical point to the next and to enter the target segments, this is","section":"§2.5, Proposition 2.19; also §2.7, Claim 2.27"},{"comment":"The composition map s ↦ (c_+, c_−) ↦ (ℓ,u) is printed with ℓ = (c_+ − c_−)/2 and u = (c_+ − c_−)/2. These two formulas are identical. Since c_± = u ± ℓ, the second coordinate must be u = (c_+ + c_−)/2. As printed, the map is degenerate and is not a linear isomorphism, which contradicts the sentence that follows ('If U = R² and the above map is the identity'). This is evidently a typo, but it needs correction and the surrounding proof should be checked against the corrected formula.","section":"Theorem 1.11, displayed map after it"}],"minor_comments":[{"comment":"The section title refers to 'Statement 3) of Theorem 1.2', but Theorem 1.2 has only two numbered statements; the second statement concerns constrictions. The numbering in the heading should be adjusted.","section":"§1.4 heading"},{"comment":"In the proof of Step 5 and the following paragraph, the text cites 'Proposition 2.14, Statement (vi)', but Proposition 2.14 has only Statements (i)–(v). The intended reference is almost certainly Statement (v), which gives the location of x_−(a) on the upper side of Q_a.","section":"§2.7, Step 5 and Proposition 2.14"},{"comment":"There are small typos: 'RJS model' should be 'RSJ model' in the introduction; 'atunnelling' has a missing space in the abstract; and 'rotatiton' appears in Remark 1.12. These do not affect the mathematics.","section":"Throughout"},{"comment":"Since Proposition 2.19 is essential but not proved, the references should be made more precise: give the exact theorem/proposition numbers, page numbers, and, if possible, state the hypotheses under which the exponential contraction is uniform in parameters. The current citation to a thesis and a paper is too vague for a load-bearing lemma.","section":"Reference [21] and [28]"}],"recommendation":"major_revision","confidential_remarks":"For the editor: the central result is likely correct, and the proof is detailed except for the flowbox lemma. The main decision hinges on whether the author can either prove Proposition 2.19 in the required generality or supply a precise reference that covers the class of systems (1.13) with conditions (i)–(vi). I would not reject the paper, but it should not be accepted in its current form with this external-support gap."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a real result. Glutsyuk proves that in the RSJ model, with B=ℓω and A=1+uω, the phase-lock area L_r(ω) converges to the parquet L^0_r, and he states a generalization for a class of slow-fast systems on T^2. The parquet was conjectured from numerics; this paper turns it into a theorem. That is genuinely new and significant. The proof is long and detailed, the Riccati rescaling and pole-counting via Hermite polynomials are elegant, and I see no fitted parameters or post-hoc exclusions.\n\nThe main soft spot is Proposition 2.19, the stable flowbox lemma. It is load-bearing: it carries orbits from the initial segment to the first Riccati neighborhood and again along the stable graph. The proposition is not proved; it is dismissed as classical and cited to [28]. Remark 1.15 even says [28] covers only a subfamily, so the citation is not obviously sufficient in the generality of Theorem 1.11. If uniform exponential contraction fails on compact parameter rectangles, the square-filling argument collapses. This is an external-support gap, not an internal inconsistency, and it is addressable, but the author needs to prove the lemma or give a precise reference that covers the full class.\n\nThere is also a clear typo in Theorem 1.11: the printed formula for u equals the formula for ℓ. It should be u=(c_+ + c_-)/2. Minor, but it will confuse readers.\n\nThe RSJ theorem itself looks credible, and the constriction-vertex argument in Section 2.8 is a nice piece of work. I did not check every estimate in the 46-page proof, but the structure is coherent and the dependence on external results is concentrated in that one flowbox lemma.\n\nWho is this for: people working on slow-fast systems, Arnold tongues, and Josephson junction models. It deserves a serious referee. I would send it out rather than desk reject, with the instruction that Proposition 2.19 be made precise. If that gets fixed, I would take the RSJ theorem as established; the general theorem is slightly more conditional but plausible.","headline":"Solid new theorem on Josephson phase-lock parquet, with one load-bearing cited flowbox lemma that should be proved or pinned down before publication.","tokens_in":37825,"tokens_out":2736,"would_cite":true,"duration_ms":27183,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["34E15","37C55","34C15"],"pacs":[],"model":"deepseek-v4-flash","headline":"In a special slow-fast limit, the phase-lock areas of the Josephson junction model converge to a parquet of unit squares and strips.","keywords":["phase-lock areas","Arnold tongues","Josephson junction","RSJ model","slow-fast systems","rotation number","Riccati equation","parquet"],"falsifier":"Take ω=0.05 and parameters (ℓ,u)=(1,2), which lies inside the r=1 square of the predicted parquet for the RSJ model; numerically integrate the equation dθ/dτ=(1/ω)(cosθ+ω+(1+2ω)cosτ) and compute the rotation number over many periods — the theorem says it equals 1. If it does not, or if the constriction of the phase-lock boundary near (1,1) does not approach that vertex as ω decreases, the convergence claim is falsified.","tokens_in":36823,"feed_emoji":"🧩","tokens_out":9032,"duration_ms":70939,"temperature":0.7,"pith_summary":"The paper establishes the first explicit asymptotic portrait of the phase-lock areas (Arnold tongues) of the overdamped Josephson junction model in the slow-fast limit where the frequency ω tends to zero and the bias parameters approach (B,A)=(0,1) at speed proportional to ω. After rescaling B=ℓω and A=1+uω, the r-th phase-lock area is proved to converge, in the Hausdorff sense, to a piecewise-linear parquet: an infinite chain of unit squares with integer vertices whose diagonals lie on the line ℓ=r, together with a downward strip (for r=0, a sector). Every vertex of this limit parquet is shown to be the limit of constrictions, the crossing points of the two boundary curves of the phase-lock area. The same result is extended to a general class of slow-fast systems on the two-torus with two Morse critical points, where the two parameters are replaced by the two constants of an associated Riccati equation. If correct, this gives a complete geometric description of the asymptotics of Arnold tongues in this regime, resolving a long-standing open question.","feed_headline":"Josephson phase-lock areas turn into a square parquet","feed_subtitle":"Rescaled near (0,1), the tongues stack as unit squares with constrictions at every vertex as ω→0.","key_machinery":"Two mechanisms carry the proof. (1) Near each critical point of the slow curve {cosθ+cosτ=0}, a local rescaling turns the system into a Riccati equation dx/dy = x² − y² + c with c=u±ℓ. Its stable solution x_−(y) has exactly m simple real poles when c∈(2m−1,2m+1), and a heteroclinic connection exactly when c=2m+1 (mirroring the harmonic-oscillator eigenvalues). The poles split the graph of x_− into arcs that correspond one-to-one to the squares of the limit parquet. (2) The classical stable-flowbox lemma (cited, not proved here) says a segment transverse to the slow curve is carried along it in an O(ω)-close ribbon of exponentially small width; this converts the local Riccati dynamics into a","core_discovery":"The central claim: in the scaling B=ℓω, A=1+uω, as ω→0, the phase-lock area L_r(ω) — the parameter set where the rotation number equals r — converges to the closure of L⁰_r = ⋃_{k,k+r≥0} {u+ℓ ∈ (2(k+r)−1, 2(k+r)+1), u−ℓ ∈ (2k−1, 2k+1)}, with the interval at index 0 equal to (−∞,1). In the (ℓ,u) plane this is a chain of unit squares with integer vertices on the line ℓ=r, plus a downward strip (a sector when r=0). The vertices on ℓ=r are exactly the limits of constrictions of L_r(ω); and the same parquet, with u±ℓ replaced by general constants c_±, is proved for a class of slow-fast systems on the two-torus with two Morse critical points.","pith_inferences":["If the convergence is as strong as the Hausdorff statement, the Arnold-tongue structure in this scaling limit is purely combinatorial: the integer lattice points on ℓ=r enumerate the constriction limits, suggesting a topological index (difference of crossing counts) could label tongues in more general junctions.","The appearance of the harmonic-oscillator eigenvalues 2m+1 in the heteroclinic condition points to a spectral mechanism behind the parquet; similar square tilings may arise in other slow-fast limits reducible to an oscillator equation, such as certain Josephson arrays or modulated planar rotors.","A natural extension is to test numerically whether the parquet persists for other scaling rates, e.g., (B,A)−(0,1) of order ω^p with p≠1; the theorem does not address those rates and the limit might differ.","The paper treats the special point (0,1); applying the same rescaling near other points of the (B,A) plane where the slow curve has different topology could yield different tilings, and the general theorem may be the first step toward a classification."],"forward_implications":["The asymptotic shape of every Arnold tongue in the (ℓ,u) plane is universal: for rotation number r it is always the same chain of unit squares plus a strip/sector, independent of the particular trigonometric forcing once the two parameters u±ℓ are fixed.","The constrictions of the phase-lock areas accumulate exactly at the integer vertices of the limit parquet on ℓ=r, predicting where boundary crossings converge.","The boundaries between neighboring phase-lock areas collapse to the piecewise-linear lines u±ℓ = 2k+1, so in the limit adjacent tongues meet along straight segments.","The general theorem extends the same conclusion to any slow-fast system on the two-torus with two Morse critical points satisfying the stated conditions, with the parquet coordinatized by the two Riccati constants c_±.","In the limit, the rotation number is the difference m−k of the half-interval indices occupied by the two parameters c_+ and c_−."],"fun_headline_variants":["Phase-lock areas become square parquet in slow-fast limit","Josephson phase-lock areas converge to square parquet","Slow-fast limit turns phase-lock tongues into square parquet","Phase-lock area portrait converges to square parquet in limit","Square parquet emerges from Josephson phase-lock areas"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The proof relies on a classical slow-fast lemma, cited without proof, asserting that a segment transverse to the slow curve is carried along it in an exponentially thin ribbon at distance O(ω); if that uniform thinness fails, the filling of the limit squares collapses.","fun_headline_variants_meta":{"raw":{"variants":["Phase-lock areas become square parquet in slow-fast limit","Josephson phase-lock areas converge to square parquet","Slow-fast limit turns phase-lock tongues into square parquet","Phase-lock area portrait converges to square parquet in limit","Square parquet emerges from Josephson phase-lock areas"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000808,"raw_usage":{"total_tokens":3500,"prompt_tokens":981,"completion_tokens":2519,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":725,"completion_tokens_details":{"reasoning_tokens":2435}},"tokens_in":725,"tokens_out":2519,"duration_ms":17346,"temperature":1.0,"reasoning_tokens":2435,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T00:37:17.268861+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take ω=0.05 and parameters (ℓ,u)=(1,2), which lies inside the r=1 square of the predicted parquet for the RSJ model; numerically integrate the equation dθ/dτ=(1/ω)(cosθ+ω+(1+2ω)cosτ) and compute the rotation number over many periods — the theorem says it equals 1. If it does not, or if the constriction of the phase-lock boundary near (1,1) does not approach that vertex as ω decreases, the convergence claim is falsified.","supporting_citations":[],"review_version":1}