{"id":"070b01eb-85bc-4538-a7c4-dc1ca9afb8d8","arxiv_id":"2505.11727","paper_version":2,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Thermal antiferromagnetic fluctuations beyond Eliashberg theory reproduce the EDC pseudogap, the gossamer Fermi surface, and the hot-spot maximum of the superconducting gap in electron-doped cuprates.","lead":"This paper shows that thermal magnetic fluctuations, treated beyond the standard Eliashberg approximation, can create the pseudogap and the 'gossamer' Fermi surface seen in electron-doped cuprates. The result explains why the superconducting gap is largest where the normal-state spectral weight is most depleted, resolving a paradox reported in the ARPES data.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The weak-pseudogap premise is not independently established: the chosen λ_th=0.85 sits near the strong-coupling boundary, where omitted O(λ_th^2) vertex and two-loop self-energy corrections are of order one and could alter the one-loop EDC/MDC predictions and the thermal near-cancellation.","rationale":"The paper has a clear, analytically explicit construction: the one-loop self-energy Eq. (1) gives Eq. (2), the threshold λ_c≈0.47 follows from expanding the hot-spot spectral function, and the EDC/MDC dichotomy is a genuine qualitative prediction. The comparison with Xu et al. is favorable for several cuts, and the authors honestly note where the fit degrades (cut 2 discrepancies, cut 4 missing peaks). The doping evolution in Appendix B is a useful falsifiable prediction. The load-bearing weakness is upstream: the entire computation is trusted only if x=0.15 is in the weak-PG regime with λ_th∈(λ_c,1], but λ_th is not independently measured. The chosen value 0.85 is close to the strong-PG boundary, and at that value the perturbative control is questionable because the next-order corrections are O(λ_th^2). This does not invalidate the scenario, but it makes the quantitative EDC/MDC comparison and the near-cancellation claim conditional, especially since the pairing-vertex estimate uses the same truncated self-energy. The reader's verdict of CONDITIONAL is appropriate; my concern is essentially the same as the reader's weakest assumption, so no verdict change is needed.","tokens_in":17248,"tokens_out":23134,"duration_ms":231528,"concrete_test":"Recompute the spectral function A(k,ω) at λ_th=0.85, ξ=10a, and the paper's band parameters with the two-loop self-energy and the leading vertex correction included (or with fRG at the same parameters), and compare EDC peak positions and the ω=0 MDC peak with Eq. (2); if either shifts by more than about 20%, the one-loop weak-pseudogap result is not quantitatively controlled.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central scenario requires NCCO at x=0.15 to be in the weak pseudogap regime, where the bare one-loop self-energy of Eq. (1) is accurate. The paper's only support is the inference in Sec. II that Δ_PG<100 meV is smaller than v_F ξ^{-1}≈100 meV, and the choice λ_th=0.85 \"for definiteness\" inside (λ_c≈0.47,1]. These do not independently establish the regime: λ_th is not derived from measured ξ(T), T, and the spin-fermion coupling; it is a fit parameter. The same data are then used to validate the spectral functions, so the comparison is partly circular. More importantly, the expansion parameter is λ_th itself: at λ_th=0.85 the neglected two-loop self-energy and vertex corrections are O(λ_th^2)≈0.7, the same order as the one-loop term. Because the paper argues self-energy and vertex insertions must be treated on equal footing, this is not a controlled perturbative truncation. The sensitivity is visible even within the paper: raising λ_th to 0.95 qualitatively restores EDC peaks in cut 4 that are absent at 0.85, and lowering it below 0.47 removes the pseudogap. Thus the EDC/MDC predictions, and the thermal near-cancellation in the pairing vertex that uses the same Σ_th, are conditional on an unverified parameter choice and on the truncation error being negligible.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper argues that the pseudogap in electron-doped cuprates at x=0.15 is a thermal precursor to antiferromagnetism, with no Fermi-surface reconstruction. Building on earlier work by Ye et al., the authors consider the weak-pseudogap regime and compute the one-loop thermal self-energy from static antiferromagnetic fluctuations using the bare fermion Green's function, explicitly going beyond the Eliashberg approximation. From the resulting spectral function they derive a threshold thermal coupling lambda_th,c ≈ 0.47 for pseudogap formation in EDCs, and they show that MDCs at zero frequency peak at the free-fermion Fermi surface, producing a 'gossamer' Fermi surface. They compare EDC and MDC spectra along six momentum cuts with the ARPES data of Xu et al., including integrated spectral weight and the doping evolution of a low-energy peak. Finally, they analyze the linearized gap equation and argue that thermal fluctuations nearly cancel in the pairing vertex, so that the d-wave gap retains its maximum at the hot spot despite the maximal normal-state spectral-weight depletion there.","tokens_in":17607,"tokens_out":8460,"duration_ms":79611,"significance":"The central qualitative claim—that the EDC/MDC dichotomy observed by Xu et al. can arise from thermal magnetic fluctuations without Fermi-surface reconstruction—is attractive and nontrivial. The analytic derivation of the spectral function (Eqs. (1)-(3)) is clean, and the explicit threshold lambda_th,c is a genuine model prediction rather than a fit. The contrast with the self-consistent one-loop approximation in Appendix A, where no pseudogap appears, clearly demonstrates the importance of using the bare Green's function. The theory also makes falsifiable predictions, such as the doping dependence of the low-energy EDC peak in Appendix B and the absence of a second peak in cuts 1-3 in Appendix C. If the near-cancellation of thermal fluctuations in the gap equation holds quantitatively, the paper offers a natural resolution of the 'hot-spot paradox'. The significance is limited by the fact that the quantitative comparison with ARPES depends on parameters (lambda_th, xi) that are not independently derived, and by the lack of a controlled small parameter at lambda_th=0.85.","major_comments":[{"comment":"The one-loop perturbative self-energy with the bare Green's function is used at lambda_th=0.85, but the expansion parameter is lambda_th itself, and two-loop self-energy and vertex corrections are of order lambda_th^2 ≈ 0.72, comparable to the retained one-loop term. The threshold lambda_th,c ≈ 0.47 already involves the O(lambda^2) term in the denominator of Eq. (3), so the truncation is not systematically controlled. Please provide a quantitative estimate of the omitted two-loop and vertex contributions at lambda_th=0.85, or demonstrate that the qualitative results are unchanged for a smaller lambda_th, e.g., 0.6, where the expansion is better controlled.","section":"Sec. II, Eqs. (1)-(3)"},{"comment":"The near-cancellation of thermal fluctuations in the gap equation is established only by a scaling estimate for k = k_h (J ~ 1/(k_h xi)^2) and by a continuity argument for other momenta. No numerical evaluation of J_k or J*_k is provided, and the estimate relies on assumptions (i) and (ii) that are not derived in this paper. Because the claim that the d-wave gap maximum remains at the hot spot is a central result, please evaluate J_k along the entire Fermi surface, or at least for representative momenta away from k_h, to confirm the smallness.","section":"Sec. IV, Eqs. (7)-(8)"},{"comment":"The parameters lambda_th=0.85 and xi=10a are chosen 'for definiteness' rather than derived from independent measurements of T, xi(T), and the spin-fermion coupling constant. The comparison with ARPES is therefore partly circular, and its sensitivity is visible in Fig. 3(d), where increasing lambda_th to 0.95 restores EDC peaks in cut 4 that are absent at 0.85. Please provide a robustness study over the allowed lambda_th window and, if possible, derive lambda_th from the measured xi(T) and T, or explicitly state which conclusions are parameter-independent.","section":"Sec. II and Figs. 3-5"},{"comment":"The assignment of NCCO at x=0.15 to the weak-pseudogap regime is inferred from the very data the theory aims to explain, and the scale separation is marginal: with xi=10a and the stated v_F, the criterion Delta_PG < v_F xi^{-1} is satisfied only by a factor of order one, not by a clear separation of scales. Please justify that the bare-Green's-function one-loop approximation is valid at this point in parameter space, for example by comparing with the numerical results of the cited earlier works for comparable parameters, or by quantifying the size of the neglected terms.","section":"Sec. II"}],"minor_comments":[{"comment":"The expression for lambda_th,c is missing a closing parenthesis: it should read lambda_th,c = 1/(1 + pi/(2 sqrt(2))) ≈ 0.47.","section":"Eq. (3) and surrounding text"},{"comment":"The prefactor in Eq. (4) is garbled; '3¯g vFξ1T' appears to be an unreadable rendering of 3 gbar T/(v_F xi^{-1}) or similar. Please fix the typesetting.","section":"Eq. (4)"},{"comment":"There are typographical slips such as 'e,g.,' instead of 'e.g.,' (pages 2 and 13) and 'linear odder in ω' instead of 'linear order in ω' (Appendix C).","section":"Throughout"},{"comment":"The text refers to Fig. 7a as the theoretical EDC spectrum, but the caption lists (a) as simply 'EDC spectrum of cut 6'; please clarify the correspondence between the panels and the theory/experiment labels.","section":"Appendix B, Fig. 7 caption"},{"comment":"The abstract states that EDC peaks occur 'at all momenta', while Fig. 3(d) shows that for lambda_th=0.85 the negative-energy EDC peak is absent for a range of momenta in cut 4. Please qualify the abstract or clarify that the positive-energy peak is meant.","section":"Abstract and Fig. 3(d)"}],"recommendation":"major_revision","confidential_remarks":"The paper is a serious candidate for publication after revision. The main concerns are that the quantitative ARPES comparison is partly fitted (lambda_th, xi) and that the gap-equation near-cancellation rests on scaling estimates rather than a controlled calculation. These are fixable in revision. The central scenario is plausible and the explicit threshold lambda_th,c plus the EDC/MDC dichotomy are valuable contributions. I would not reject; the revision should focus on strengthening the parameter justification and the gap-equation analysis."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper's real contribution is the beyond-Eliashberg one-loop treatment of thermal AFM fluctuations, and the explicit demonstration that EDC and MDC behave differently: EDC peaks at finite frequency (pseudogap) while MDC disperses to the free-fermion Fermi surface (gossamer). That dichotomy is clean and follows from Eq. (2) with a closed-form threshold lambda_th,c ~ 0.47. The comparison with Xu et al. is detailed—hot-spot maximum of the d-wave gap, non-monotonic EDC dispersion, the cut-6 features—and it is the strongest part of the paper. The SC section is plausible but less solid.\n\nThe soft spots are real, and the stress-test note is on target. The expansion parameter is lambda_th itself: at the chosen 0.85, the neglected O(lambda^2) corrections are of order one. The paper rightly insists that self-energy and vertex insertions must be treated on equal footing, which means a bare one-loop truncation is not controlled at that coupling. The sensitivity is visible in the paper's own Fig. 3d: at 0.95 the cut-4 EDC peaks reappear. Also, lambda_th and xi are chosen to reproduce the very ARPES features being explained, so part of the agreement is parameterized. The superconductivity near-cancellation is argued by scaling and a 'by continuity' expectation, not by a numerical solution of the gap equation; that is a gap, though a fillable one.\n\nNone of this kills the paper. The threshold and the EDC/MDC dichotomy are robust structural properties of the one-loop model, not fit. The data comparison is extensive enough that even a skeptical reader learns something. I agree with the conditional verdict. A serious referee should ask for two things: a numerical solution of the full nonlinear gap equation with and without the thermal piece, and a systematic estimate of the two-loop and vertex corrections at lambda=0.85, or an argument that the effective coupling is smaller. This is a paper for the cuprate theory community, and it deserves the referee time.","headline":"A clean EDC/MDC dichotomy from one-loop thermal fluctuations, but the weak-coupling regime is handpicked and the SC cancellation is argued rather than shown.","tokens_in":18142,"tokens_out":2021,"would_cite":true,"duration_ms":22481,"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 the pseudogap in electron-doped cuprates is a thermal precursor to antiferromagnetism: in the weak pseudogap regime, a one-loop self-energy from thermal magnetic fluctuations reproduces the ARPES EDC and MDC spectra…","keywords":["pseudogap","electron-doped cuprates","thermal magnetic fluctuations","ARPES spectral function","gossamer Fermi surface","hot spot","d-wave superconductivity","spin-fermion model"],"falsifier":"Measure the hot-spot EDC with resolution sufficient to resolve $\\omega=0$: in the weak regime the theory requires a local minimum at $\\omega=0$ (two peaks at finite $\\pm\\omega$), whereas a reconstructed state or a self-consistent one-loop calculation gives a single peak at $\\omega=0$; likewise, an MDC peak at $\\omega=0$ displaced from the free-fermion $\\mathbf{k}_F$ on the same cuts would falsify the claim.","tokens_in":17024,"feed_emoji":"🧲","tokens_out":13441,"duration_ms":112565,"temperature":0.7,"pith_summary":"The paper argues that the pseudogap in electron-doped cuprates is not a signature of a hidden ordered state or a reconstructed Fermi surface, but a thermal precursor to antiferromagnetism: the same thermal magnetic fluctuations that destroy long-range $(\\pi,\\pi)$ order at $T_N$ transfer spectral weight away from zero frequency when the system is just above $T_N$. Using a perturbative one-loop self-energy built from the static antiferromagnetic propagator and a bare fermion propagator, the authors show that energy distribution curves (EDC, intensity at fixed momentum versus frequency) acquire peaks at finite frequency at all momenta, while momentum distribution curves (MDC, intensity at fixed frequency versus momentum) still peak at the free-fermion Fermi surface at zero frequency, forming what the experiment called a gossamer Fermi surface. Applied to the electron-doped cuprate NCCO at $x=0.15$, the calculation reproduces the EDC and MDC peak positions, the minimum of low-energy spectral weight at the hot spot, and the doping evolution of a low-energy EDC peak. The paper further shows that thermal fluctuations almost cancel out of the superconducting gap equation, so the $d$-wave gap keeps its maximum at the hot spot, resolving the apparent paradox in the experimental data. If correct, the pseudogap in these materials is explained without invoking Fermi-surface reconstruction.","feed_headline":"Cuprate pseudogap traced to thermal magnetism","feed_subtitle":"A one-loop thermal spin-fluctuation calculation reproduces ARPES peak shapes without Fermi-surface reconstruction.","key_machinery":"The load-bearing object is the one-loop thermal self-energy $\\Sigma_{\\mathrm{th}}(\\mathbf{k},\\omega)$, the convolution of the static antiferromagnetic susceptibility $\\chi(\\mathbf{q}+\\mathbf{Q})\\propto 1/(q^2+\\xi^{-2})$ with a bare fermion propagator, evaluated with the bare Green's function rather than a self-consistent one; the authors stress that this perturbative choice is required because for thermal fluctuations self-energy and vertex corrections must be dropped on equal footing. From this self-energy they derive a closed-form spectral function whose EDC curvature changes sign at $\\lambda_{\\mathrm{th},c}=1/(1+\\pi/(2\\sqrt{2}))\\approx 0.47$, while the corresponding MDC curvature does not. For superconductivity, the key identity is the rewrite of the pairing vertex as $\\Phi(\\mathbf{k},\\omega_m)=\\Delta(\\mathbf{k},\\omega_m)(\\omega_m+\\Sigma_{\\mathrm{th}}(\\mathbf{k},\\omega_m))/\\omega_m$, which isolates the thermal $\\Omega_m=0$ piece of the gap equation; because the ratio $\\Delta(\\mathbf{k}+\\mathbf{q}\\xi^{-1})/\\Delta(\\mathbf{k})$ is nearly unity over the relevant momentum range, the thermal correction $J\\sim 1/(k_h\\xi)^2$ is small. This machinery converts thermal magnetic fluctuations into both the pseudogap phenomenology and the preservation of the hot-spot maximum of the $d$-wave gap.","core_discovery":"The central claim is that in the weak pseudogap regime, where the pseudogap scale $\\Delta_{\\mathrm{PG}}$ is smaller than $v_F\\xi^{-1}(T)$, the fermionic spectral function obtained from the one-loop thermal self-energy already contains all the qualitative features seen in ARPES on electron-doped cuprates. At a hot spot (the Fermi momentum $\\mathbf{k}_h$ for which $\\mathbf{k}_h+\\mathbf{Q}$ is also on the Fermi surface), the EDC $A_{\\mathbf{k}_h}(\\omega)$ has two peaks at finite $\\omega=\\pm\\Delta_{\\mathrm{PG}}$ once the thermal coupling $\\lambda_{\\mathrm{th}}$ exceeds $\\lambda_{\\mathrm{th},c}\\approx 0.47$; at every momentum along the measured cuts, EDC peaks sit at finite frequency, whereas the MDC at $\\omega=0$ peaks at the free-fermion Fermi momentum $\\mathbf{k}_F$, reproducing the gossamer Fermi surface (a Fermi surface with reduced but finite spectral weight). The same self-energy, inserted into the linearized gap equation, yields a thermal correction factor $J\\sim 1/(k_h\\xi)^2$ that is small, so thermal fluctuations do not reshape the momentum dependence of the pairing gap, which retains its maximum at the hot spot. The paper asserts that this matches the experimental EDC and MDC spectra, the minimum of integrated spectral weight at the hot spot, and the doping evolution of the low-energy peak, all without Fermi-surface reconstruction.","pith_inferences":["Beyond the paper: if the thermal-precursor picture is right, the same perturbative machinery should apply to other electron-doped cuprates near their antiferromagnetic endpoint; a direct test would be to measure EDC and MDC in another compound and check that the extracted $\\lambda_{\\mathrm{th}}$ falls between the threshold and unity whenever pseudogap behavior appears.","Beyond the paper: the near-cancellation of thermal fluctuations in the gap equation implies that the pseudogap itself does not suppress superconductivity through thermal pair breaking; if so, the doping at which the pseudogap opens and the doping at which superconductivity appears need not be controlled by the same fluctuations, a distinction that could be probed by tuning the magnetic correlation","Beyond the paper: the sharpest experimental consequence is that in the weak regime the MDC peak at $\\omega=0$ should sit exactly at the free-fermion Fermi momentum on every cut; high-resolution ARPES that resolved both the EDC pseudogap and an MDC peak displaced onto a reconstructed pocket edge would rule the scenario out.","Beyond the paper: a natural extension is to replace the static thermal susceptibility by the full dynamical susceptibility and check whether finite-frequency corrections shift the EDC peak positions or the near-cancellation factor $J$; the present calculation keeps the susceptibility static."],"forward_implications":["If the scenario is correct, the pseudogap in NCCO at $x=0.15$ and nearby dopings is not evidence for a hidden order or a reconstructed Fermi surface; the spectral-weight depletion is a finite-temperature effect of incipient antiferromagnetism.","The same one-loop calculation predicts that the EDC pseudogap peak position evolves non-monotonically along a momentum cut, and that in some cuts a secondary low-energy peak appears from the product of the spectral function with the Fermi function, both matching the ARPES data.","Thermal fluctuations act like non-magnetic impurities in the pairing vertex: they suppress normal-state spectral weight but almost cancel in the gap equation, so the $d$-wave gap magnitude along the Fermi surface is governed by quantum fluctuations and is largest at the hot spot.","The theory gives a concrete doping dependence: as doping increases and the correlation length $\\xi$ shrinks, the system crosses from the weak pseudogap regime to a conventional metal, with the low-energy EDC peak moving toward zero frequency.","Because no Fermi-surface reconstruction is involved, the magnetic Brillouin zone boundary is not a locus of gap opening; the apparent folded dispersion near the zone boundary in cut 6 is reproduced as a remnant of the two-band structure of the ordered state."],"supporting_citations":[{"why":"the ARPES dataset on NCCO at $x=0.15$ whose EDC and MDC spectra, spectral-weight minimum, and gap maximum the theory reproduces.","marker":"[48]"},{"why":"the prior analysis that supplies the weak versus strong pseudogap classification and the one-loop thermal self-energy expression used in the calculation.","marker":"[33]"},{"why":"the earlier thermal-fluctuation approach to the pseudogap, including the self-consistent one-loop result that shows no pseudogap and motivates using the bare propagator.","marker":"[26]"},{"why":"the spin-fluctuation result that without thermal fluctuations the $d$-wave gap is largest at the hot spot, which the near-cancellation preserves.","marker":"[53]"},{"why":"the analytical self-consistent one-loop self-energy reproduced in Appendix A to show that no pseudogap appears in that approximation.","marker":"[56]"},{"why":"the follow-up ARPES study of the doping evolution of the low-energy EDC peak, compared in Appendix B.","marker":"[52]"},{"why":"the theorem on impurity-insensitive pairing, cited as the physical analogue for why thermal fluctuations nearly cancel in the gap equation.","marker":"[69]"}],"fun_headline_variants":["Thermal magnetism sets the cuprate pseudogap scale","Cuprate pseudogap: a thermal spin-wave precursor","Gossamer Fermi surface from thermal fluctuations","One-loop thermal self-energy reproduces ARPES spectra","Pseudogap in electron-doped cuprates is thermal magnetism"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The calculation rests on the assumption that NCCO at $x=0.15$ is in the weak pseudogap regime, with pseudogap energy smaller than $v_F\\xi^{-1}(T)$, so the one-loop self-energy with bare fermions is valid and vertex corrections can be neglected; this regime assignment is inferred from the very data the paper seeks to explain.","fun_headline_variants_meta":{"raw":{"variants":["Thermal magnetism sets the cuprate pseudogap scale","Cuprate pseudogap: a thermal spin-wave precursor","Gossamer Fermi surface from thermal fluctuations","One-loop thermal self-energy reproduces ARPES spectra","Pseudogap in electron-doped cuprates is thermal magnetism"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000268,"raw_usage":{"total_tokens":1657,"prompt_tokens":1020,"completion_tokens":637,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":636,"completion_tokens_details":{"reasoning_tokens":556}},"tokens_in":636,"tokens_out":637,"duration_ms":6595,"temperature":1.0,"reasoning_tokens":556,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T20:49:31.131955+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the hot-spot EDC with resolution sufficient to resolve $\\omega=0$: in the weak regime the theory requires a local minimum at $\\omega=0$ (two peaks at finite $\\pm\\omega$), whereas a reconstructed state or a self-consistent one-loop calculation gives a single peak at $\\omega=0$; likewise, an MDC peak at $\\omega=0$ displaced from the free-fermion $\\mathbf{k}_F$ on the same cuts would falsify the claim.","supporting_citations":[{"cited_title":"Both theory and experiment show that the spectral weight reaches a minimum at the hot spot","cited_arxiv_id":null,"evidence_quote":"the ARPES dataset on NCCO at $x=0.15$ whose EDC and MDC spectra, spectral-weight minimum, and gap maximum the theory reproduces."},{"cited_title":"Grossman and E","cited_arxiv_id":null,"evidence_quote":"the prior analysis that supplies the weak versus strong pseudogap classification and the one-loop thermal self-energy expression used in the calculation."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"the earlier thermal-fluctuation approach to the pseudogap, including the self-consistent one-loop result that shows no pseudogap and motivates using the bare propagator."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"the spin-fluctuation result that without thermal fluctuations the $d$-wave gap is largest at the hot spot, which the near-cancellation preserves."},{"cited_title":"Fujimoto, Pseudogap phenomena in the BCS pairing model, Journal of the Physical Society of Japan 71, 1230 (2002); Y","cited_arxiv_id":null,"evidence_quote":"the analytical self-consistent one-loop self-energy reproduced in Appendix A to show that no pseudogap appears in that approximation."},{"cited_title":"Horio, K","cited_arxiv_id":null,"evidence_quote":"the follow-up ARPES study of the doping evolution of the low-energy EDC peak, compared in Appendix B."},{"cited_title":"Wang, Solvable strong-coupling quantum-dot model with a non-fermi-liquid pairing tran- sition, Phys","cited_arxiv_id":null,"evidence_quote":"the theorem on impurity-insensitive pairing, cited as the physical analogue for why thermal fluctuations nearly cancel in the gap equation."}],"review_version":1}