{"id":"3e3ec412-800e-4745-867b-a11f98b24c72","arxiv_id":"2412.03698","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"The dynamical pairing susceptibility in a quantum-critical metal remains finite and positive exactly at the superconducting instability for a constant source, then becomes a non-unique function of a free parameter below it.","lead":"This paper calculates the pairing susceptibility in a quantum-critical metal and finds that it does not diverge at the onset of superconductivity for a constant pairing source, unlike in BCS theory. Below the onset, the susceptibility becomes multi-valued, which the authors interpret as evidence for a multi-critical superconducting transition.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Non-divergence at Ncr hinges on a piecewise-local approximation of the pairing kernel that is only tested at γ=0.5; at the exact Ncr of Eq.(3) the approximation may fail.","rationale":"The reader's weakest-assumption is precisely the load-bearing point. The full analytical machinery—the H±b solutions, the C(b)=0 selection, and the η-family of Eq.(36)—is built on Eq.(7), not on the original nonlocal equation. The paper itself labels the reduction as valid for small γ, yet the conclusions are stated for all 0<γ<1, and Eq.(3) shows that the exact Ncr is not the small-γ expression used in Eq.(8) except near γ=0. The absence of numerical verification at other γ and near Ncr is not a cosmetic gap: the non-divergence at Ncr is a beyond-leading-logarithm effect, precisely the regime in which the piecewise-local replacement of the kernel is least controlled. The paper's assertion that solutions of Eq.(4) and Eq.(9) 'almost coincide for all γ<1' is unaccompanied by data, and Fig.4 covers only one γ and one N safely above Ncr. A direct numerical solution of Eq.(4) across γ would either validate the approximation or require abandoning the central claim. Since the concern is testable and currently unresolved, the reader's CONDITIONAL verdict is the appropriate one; my analysis does not move it.","tokens_in":20903,"tokens_out":10630,"duration_ms":111171,"concrete_test":"Solve the original integral equation (4) directly by iteration for constant Φ0 at γ=0.8 and γ=0.95, using the exact Ncr from Eq.(3). For N just above Ncr, extract χpp(z)=Φ(z)/Φ0 at fixed z=ω^γ and check whether it remains finite as N→Ncr^+, as Eq.(30) predicts, or diverges as 1/√(N−Ncr). Repeat for N just below Ncr and compare the numerically obtained ratio with Eq.(36) for any η; if no η fits, or if the iteration diverges in a way incompatible with Eq.(36), the local approximation is not controlled and the central claim fails.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—that χpp stays finite and positive at the T=0 pairing threshold for a constant source—is derived not from the original nonlocal gap equation (4) but from the local approximation (6)/(7), obtained by replacing |ω_m−ω'_m|^γ with |ω_m|^γ or |ω'_m|^γ depending on which frequency is larger. The text introduces this reduction as valid 'for small γ' and then uses it for all 0<γ<1. The threshold in Eq.(8), Ncr=4(1−γ)/γ, is the small-γ form of the exact Ncr in Eq.(3); for γ=0.8, Eq.(3) gives Ncr≈1.8 while Eq.(8) gives 1.0, and for γ→1 the discrepancy grows rapidly. The only direct numerical comparison between Eq.(4) and Eq.(7) is Fig.4, at γ=0.5 and N=6.5, i.e., far from criticality and at one γ. The statement that the solutions 'almost coincide for all γ<1' is not documented. If the local reduction changes the location or nature of the threshold, the non-divergence at Ncr and the η-dependent susceptibility below Ncr may be artifacts of Eq.(7). The argument that the H_{−b} component is not generated also relies on small-z and large-z asymptotics plus a normalizability criterion, not on solving the original equation; if the full kernel generates H_{−b}, the unique χpp above Ncr is lost.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper analyzes the T=0 dynamical pairing susceptibility χpp(ωm) in a quantum-critical metal described by the γ-model, defined as the ratio of the fully dressed infinitesimal pairing vertex Φ(ωm) to a bare source Φ0(ωm). The main results are that, for a constant source, χpp does not diverge when the pairing interaction strength N approaches the critical value Ncr from above, and remains finite and positive at N=Ncr; below Ncr it becomes a function of a continuous free parameter η, is negative for some range of η, and diverges at η=π/2. For a special class of frequency-dependent sources, χpp diverges at N→Ncr, while for generic frequency-dependent sources it remains regular. The authors interpret this behavior as evidence of a multi-critical superconducting transition, with an infinite set of pairing states emerging simultaneously below Ncr. The derivations are based on a local approximation to the nonlocal gap equation, validated numerically only for γ=0.5. The paper is clearly written and the interpretation is physically motivated, but the central claims have not yet been tied robustly to the original nonlocal equation.","tokens_in":21293,"tokens_out":5627,"duration_ms":61111,"significance":"If the central claims hold, the paper identifies a genuinely unconventional property of pairing in quantum-critical metals: the linear response to a pairing source can remain non-singular at the pairing threshold, and below threshold the susceptibility is non-unique, reflecting a multi-critical point. This would extend prior work by Abanov and Chubukov on the γ-model and provides a concrete, falsifiable distinction between BCS and quantum-critical pairing. The main strengths are the explicit analytical construction of the solution space of the approximate integral equation, the iterative numerical checks for γ=0.5, and the physical interpretation via normalizability and the analogy with the heavy-atom Klein-Gordon problem. The results are, however, derived from a piecewise-local replacement of the kernel |ωm−ω′m|γ, and the paper's own numerical verification is limited to a single value of γ far from criticality. Because the threshold and the non-divergence at Ncr are the central claims, the validity of this approximation across the full parameter range is the principal barrier to accepting the results as statements about the original model.","major_comments":[{"comment":"The central results are derived not from the nonlocal gap equation (4) but from the piecewise-local approximation (6), leading to Eq. (7) and the differential equation (9), with Ncr replaced by the small-γ form Ncr=4(1−γ)/γ in Eq. (8). The paper states that the solutions of (4) and (9) 'almost coincide for all γ<1', but the only numerical evidence is Fig. 4, computed at γ=0.5 and N=6.5, which is far above Ncr. For γ=0.8, the exact expression (3) gives Ncr≈1.8 while Eq. (8) gives 1.0; for γ→1 the discrepancy grows further. If the local reduction changes the location or nature of the threshold, the claim that χpp remains finite at the true Ncr, and the η-dependent family below Ncr, may be artifacts of Eq. (7). I request a systematic numerical comparison of the iterative solutions of (4) and (6)/(9) for several γ values, including γ close to 1, and for N close to Ncr on both sides of the transition, together with a demonstration that the qualitative behavior is unchanged.","section":"§II.A, Eqs. (4)-(9)"},{"comment":"The uniqueness of χpp above Ncr rests on the claim C(b)=0 in Eq. (19). The justification is threefold: the N→∞ limit, the small-z and large-z iterative expansions, and a normalizability criterion. The asymptotics are checked only at small and large z, and the text itself concludes with 'It is then natural to assume' that H−b(z) is not generated for any z; this is not a proof for intermediate frequencies. Moreover, footnote 61 states that the norm of both H_b and H_{−b} diverges logarithmically at z→∞ unless a convergence factor is introduced, so the norm argument in Sec. V is not decisive in the δ→0 limit. Since any admixture of H_{−b} would change χpp and could introduce a divergence at Ncr, this step is load-bearing. The authors should either prove that the iterative solution converges to H_b for all z in a suitable norm for N>Ncr, or specify and justify an additional regularity condition that selects C(b)=0 and verify it against the original integral equation (4).","section":"§III.B, Eqs. (19)-(30) and §V"},{"comment":"Below Ncr, the pairing susceptibility is presented as a function of a free parameter η with no physical selection rule. For a fixed constant source Φ0, different η give positive, negative, or divergent χpp. The paper interprets this as multi-criticality, but it is also possible that the non-uniqueness is an artifact of solving the linearized equation without specifying how the infinitesimal source is switched on or how the limit δ→0 in the convergence factor is taken. To make the claim falsifiable, the authors should state whether η labels different stationary states of the system, and how a particular η is selected in a concrete measurement or in the limiting procedure from finite T. Without such a criterion, the statement that 'the susceptibility becomes non-unique' is not a well-defined prediction about a response function.","section":"§III.C, Eqs. (35)-(38)"},{"comment":"The divergence of χpp at N→Ncr for the special family of sources (39) is derived in the approximate model and in the small-b limit, where the prefactor 1/√(1−λ) appears. Given the concern in the first major comment, this divergence and the C0=0/C0≠0 criterion below Eq. (47) should be checked against the original integral equation (4). If the local approximation shifts the critical coupling or changes the structure of the zero mode, the exponent of the divergence and the condition for its appearance may not survive.","section":"§IV, Eqs. (43)-(44)"}],"minor_comments":[{"comment":"Typographical errors: 'we immediately fund that' should read 'we immediately find that', and the fragment 'because use' appears in the sentence introducing Eq. (15); these should be corrected.","section":"§III.B, after Eq. (18)"},{"comment":"The caption of Fig. 4 should state explicitly which curve corresponds to the original integral equation (4) and which to the approximate equation (6); the text does not consistently describe the blue and red lines, and the axes labels should be defined.","section":"Fig. 4"},{"comment":"The substitution z=ω_m^γ is used throughout, but the positive-frequency branch and the symmetry of Φ(ω_m) are not stated explicitly; please specify that the equations are written for ω_m>0 and indicate how the results are continued to negative frequency.","section":"§II.A, after Eq. (7)"},{"comment":"The notation for the imaginary part of b is inconsistent: Eq. (34) defines ˜b, but Eq. (38) uses both '¯b' and 'π2' in the printed text; the expressions should use a single notation, e.g., π^2 \tilde b |tan η|, and the constant c should be defined in the same notation.","section":"§III.C, Eqs. (37)-(38)"},{"comment":"The phrase 'at N = Ncr − 0' is used without definition; a brief explanation of the convention (approach from above or below) would improve readability.","section":"Introduction, paragraph after Eq. (1)"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a real result, not a repackaging. The paper computes the dynamical pairing susceptibility chi_pp at T=0 in the gamma-model and shows that for a constant pairing source it does not diverge at Ncr; below Ncr it becomes a function of a free parameter eta. That behavior is genuinely new, and the derivation is transparent enough to check. The physical interpretation in terms of a multi-critical onset with infinitely many superconducting states is also well argued, not hand-waved.\n\nThe main soft spot is exactly where the reader and the stress-test put it: the central analysis is done in the piecewise-local approximation, Eq. (6)/(7), and the numerical comparison with the original nonlocal equation (4) is shown only for gamma = 0.5 and N = 6.5. The stress-test note is right to worry, though I think it slightly overstates the problem. The approximation is explicitly a small-gamma expansion, and the exact and approximate Ncr are close for gamma up to about 0.8. But the paper claims the solutions \"almost coincide for all gamma < 1\" without documenting that, and near gamma = 1 the approximate Ncr from Eq. (8) goes to 0 while the exact Ncr goes to 1. That is a real gap in scope, not just a missing figure. A referee should ask for numerical solutions of Eq. (4) at several gamma values, especially 0.8 and 0.9, near criticality.\n\nThe argument that the H_{-b} component is not generated by the full integral equation also deserves scrutiny. It is based on small-z and large-z asymptotics plus normalizability, not on a direct check of the nonlocal kernel. If the full kernel generates H_{-b}, the unique chi_pp above Ncr could be lost. This is the other thing I would want addressed.\n\nWhere the paper is solid: the eta-dependent susceptibility below Ncr follows mathematically from the homogeneous equation having normalizable solutions, which was established in prior work by the same group. The frequency-dependent source analysis, including the divergence for special sources and the completeness of the Phi_tilde-b functions, is a clear step beyond earlier papers. The condensation-energy/norm argument for selecting H_b is physically sensible. The citation pattern is appropriate; the heavy reliance on the authors' previous papers is legitimate because those papers provide the input solutions.\n\nThis paper deserves a serious referee, not a desk reject. I would recommend engaging, with revision requests focused on numerical verification of the local approximation across gamma and on the H_{-b} question in the full equation. If those checks pass, the qualitative story is likely right.","headline":"A genuinely new calculation of the dynamical pairing susceptibility in the gamma-model, with the main caveat that the central result rests on a piecewise-local approximation whose numerical validation is thin away from gamma = 0.5.","tokens_in":21720,"tokens_out":2590,"would_cite":true,"duration_ms":30412,"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":"The paper claims that for a constant source, the zero-temperature pairing susceptibility in a quantum-critical metal stays positive and finite at the superconducting onset, and becomes a one-parameter family below it.","keywords":["pairing susceptibility","quantum-critical metal","non-Fermi liquid","superconductivity","Eliashberg theory","gamma model","multi-critical point","dynamical pairing vertex"],"falsifier":"Solve the original non-local integral equation (4) numerically at $T=0$ with a constant source for $\\gamma$ well away from $0.5$, say $\\gamma=0.8$, and examine $\\chi_{pp}(\\omega_m)$ as $N$ approaches $N_{cr}$ from above; if the susceptibility develops a divergence or a sign change at criticality, the non-divergence and non-uniqueness claimed here do not survive the approximation.","tokens_in":20712,"feed_emoji":"⚛️","tokens_out":13036,"duration_ms":101807,"temperature":0.7,"pith_summary":"At zero temperature, the paper studies how a quantum-critical metal responds to an infinitesimally small pairing field, through the dynamical pairing susceptibility $\\chi_{pp}(\\omega_m)=\\Phi(\\omega_m)/\\Phi_0(\\omega_m)$. It claims that the BCS rule of thumb—susceptibility positive above the transition, divergent at it, negative below—does not hold for pairing out of a non-Fermi liquid. For a frequency-independent source, $\\chi_{pp}$ remains positive and finite as the pairing threshold $N=N_{cr}$ is approached from above, and only for $N<N_{cr}$ does it become a non-unique function of a free parameter $\\eta$, with a divergence at $\\eta=\\pi/2$. The paper reads this as evidence that the $T=0$ onset of superconductivity in a quantum-critical metal is a multi-critical point at which infinitely many superconducting states with different order-parameter amplitudes emerge simultaneously.","feed_headline":"Pairing susceptibility stays finite at superconductivity onset","feed_subtitle":"In a quantum-critical metal, a constant pairing probe shows no critical divergence—the onset is multi-critical.","key_machinery":"The argument runs through the reduction of the non-local integral gap equation (4) to the approximate local equation (7), replacing $|\\omega_m-\\omega'_m|^\\gamma$ by the larger of $|\\omega_m|^\\gamma$ and $|\\omega'_m|^\\gamma$. Differentiating (7) twice yields a second-order differential equation (9) whose two independent solutions are hypergeometric functions $H_b(z)$ and $H_{-b}(z)$ with $z=\\omega_m^\\gamma$ and $b=\\frac12\\sqrt{1-N_{cr}/N}$. Normalizability under the condensation-energy scalar product picks $H_b$ for $N>N_{cr}$; below $N_{cr}$, $b=i\\tilde b$ and the real response is a linear combination of the two complex-conjugate functions with a free phase parameter $\\eta$. The divergence at $\\eta=\\pi/2$ corresponds to the limiting non-normalizable solution of the linearized gap equation, whose condensation energy is logarithmically divergent and regularized only by nonlinearity.","core_discovery":"The paper's central claim is that the zero-temperature pairing susceptibility in the $\\gamma$-model of a quantum-critical metal is not the BCS susceptibility. When the external source $\\Phi_0$ is constant, $\\chi_{pp}$ is a regular, positive function for all $N\\geq N_{cr}$, including at $N=N_{cr}$; no divergence flags the pairing instability. Immediately below $N_{cr}$, the perturbative iteration series diverges and the susceptibility becomes a one-parameter family $\\chi_{pp}(z,\\eta)$ built from the complex-conjugate solutions $H_{i\\tilde b}(z)$ and $H_{-i\\tilde b}(z)$; it is negative for a range of $\\eta$ near $\\pi/2$ and diverges at $\\eta=\\pi/2$. For a frequency-dependent source, the same picture holds generically, but if the source contains the specific component $\\partial H_b/\\partial b|_{b=0}$, the susceptibility diverges as $1/\\sqrt{N_{cr}-N}$ when $N\\to N_{cr}$ from above. The paper interprets the non-unique, parameter-dependent response below $N_{cr}$ as the signature of a multi-critical onset, consistent with the previously found infinite family of topologically distinct gap solutions.","pith_inferences":["If this picture survives in the full non-local model, the pairing susceptibility in quantum-critical metals should be understood as a multi-valued response: different probes can see either a regular or a divergent susceptibility at the same critical coupling, depending on the frequency profile of the probe.","The mathematical structure—log-oscillatory non-normalizable solutions regularized by nonlinearity—is the same as in the superheavy-atom problem noted by the authors; one could test the analogy by looking for the same $\\eta$-family of responses in other symmetry-breaking problems with singular long-range kernels.","The mechanism suggests a concrete experimental signature: a static pairing probe should show a smooth, non-divergent response as a quantum-critical metal is tuned through the superconducting onset, whereas a probe with a specific frequency structure should show a $1/\\sqrt{N_{cr}-N}$ divergence."],"forward_implications":["A non-divergent $\\chi_{pp}$ above the onset does not mean the normal state is stable; the instability instead shows up as the breakdown of the iteration series at $N<N_{cr}$.","Below $N_{cr}$, the pairing susceptibility is intrinsically non-unique, so a response experiment or calculation must specify which $\\eta$ sector the probe couples to.","At $N=N_{cr}$ and constant source, $\\chi_{pp}(z)$ behaves as $z^{-1/2}$ at small frequency, a power law without any critical pole.","For frequency-dependent sources, the divergence at criticality is controlled by the overlap of the source with the zero-mode component $\\partial H_b/\\partial b|_{b=0}$; generic sources remain regular."],"supporting_citations":[{"why":"Provides the $\\gamma$-model phase diagram, the critical value $N_{cr}$, and the infinite discrete set of topologically distinct gap solutions that the susceptibility analysis builds on.","marker":"[50]"},{"why":"Justifies the model setup in which the particle-particle interaction is reduced by a factor $1/N$ relative to the particle-hole interaction.","marker":"[47]"},{"why":"Earlier comparison of coherent versus incoherent pairing that identified the logarithmic series and power-law form of $\\chi_{pp}$, the starting point of the present beyond-leading-log calculation.","marker":"[6]"},{"why":"Establishes the Yukawa-SYK framework in which the relative strength of the pairing interaction is a continuously tunable parameter, motivating treatment of $N$ as continuous.","marker":"[45]"},{"why":"Shows that a local Eliashberg-type description with local self-energy and pairing vertex is accurate for quantum-critical metals, justifying the local gap equation.","marker":"[49]"},{"why":"Supplies the earlier observation of multiple topologically distinct solutions of the pairing problem that the multi-critical interpretation of $\\chi_{pp}$ is designed to reconcile.","marker":"[55]"}],"fun_headline_variants":["Quantum-critical pairing onset: susceptibility shows no BCS divergence","Multi-critical superconductivity arises from non-BCS pairing response","Pairing susceptibility defies BCS near quantum-critical onset","Finite pairing susceptibility hints at multi-critical superconducting onset","No BCS pole: pairing susceptibility stays regular at quantum-critical onset"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing simplification is the replacement of the non-local kernel $|\\omega_m-\\omega'_m|^\\gamma$ by the piecewise-local form that turns the integral gap equation into the differential equation (9); the paper's main conclusions are derived from this approximate equation, and the direct numerical check against the original equation is shown only for $\\gamma=0.5$.","fun_headline_variants_meta":{"raw":{"variants":["Quantum-critical pairing onset: susceptibility shows no BCS divergence","Multi-critical superconductivity arises from non-BCS pairing response","Pairing susceptibility defies BCS near quantum-critical onset","Finite pairing susceptibility hints at multi-critical superconducting onset","No BCS pole: pairing susceptibility stays regular at quantum-critical onset"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000456,"raw_usage":{"total_tokens":2361,"prompt_tokens":1088,"completion_tokens":1273,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":704,"completion_tokens_details":{"reasoning_tokens":1191}},"tokens_in":704,"tokens_out":1273,"duration_ms":10667,"temperature":1.0,"reasoning_tokens":1191,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T22:10:22.290626+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Solve the original non-local integral equation (4) numerically at $T=0$ with a constant source for $\\gamma$ well away from $0.5$, say $\\gamma=0.8$, and examine $\\chi_{pp}(\\omega_m)$ as $N$ approaches $N_{cr}$ from above; if the susceptibility develops a divergence or a sign change at criticality, the non-divergence and non-uniqueness claimed here do not survive the approximation.","supporting_citations":[{"cited_title":"Abanov and A","cited_arxiv_id":null,"evidence_quote":"Provides the $\\gamma$-model phase diagram, the critical value $N_{cr}$, and the infinite discrete set of topologically distinct gap solutions that the susceptibility analysis builds on."},{"cited_title":"Raghu, G","cited_arxiv_id":null,"evidence_quote":"Justifies the model setup in which the particle-particle interaction is reduced by a factor $1/N$ relative to the particle-hole interaction."},{"cited_title":"Abanov, A","cited_arxiv_id":null,"evidence_quote":"Earlier comparison of coherent versus incoherent pairing that identified the logarithmic series and power-law form of $\\chi_{pp}$, the starting point of the present beyond-leading-log calculation."},{"cited_title":"Classen and A","cited_arxiv_id":null,"evidence_quote":"Establishes the Yukawa-SYK framework in which the relative strength of the pairing interaction is a continuously tunable parameter, motivating treatment of $N$ as continuous."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Shows that a local Eliashberg-type description with local self-energy and pairing vertex is accurate for quantum-critical metals, justifying the local gap equation."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the earlier observation of multiple topologically distinct solutions of the pairing problem that the multi-critical interpretation of $\\chi_{pp}$ is designed to reconcile."}],"review_version":1}