{"id":"a1dccc02-d8a1-4c24-85f7-718fc9cc33c1","arxiv_id":"1908.11795","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Using Gaussian pair fluctuations and analytic continuation to complex frequencies, the authors compute finite-temperature Leggett mode frequencies and damping in two-band Fermi superfluids, finding the mode survives at T_c in the BEC regime and anticrosses the pair-breaking edge away from BEC.","lead":"This paper calculates how a collective oscillation called the Leggett mode behaves in a two-band superfluid at finite temperature. The results suggest this mode should be observable in ultracold atomic gases in the BEC regime, which could guide future experiments.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Damped Leggett modes rest on an unverified continuation: Eq. (19) selects a lower-half-plane sheet, and the paper never checks that the resulting complex roots are poles of the physical retarded response rather than artifacts.","rationale":"The reader's weakest assumption is exactly the analytic continuation in Eq. (19), and I agree that it is the load-bearing point. The paper is transparent about its method, gives explicit matrix elements, and acknowledges that the uncontinued determinant has no complex roots; this makes the continuation step the only source of the damped modes. The qualitative spectral-weight plots are helpful but do not settle the sheet-selection question, because a spurious root can sit on a part of the lower half-plane that carries no spectral weight, while the true resonance is elsewhere. The proposed peak-versus-pole comparison is a direct, computationally cheap falsification test and can be run independently from the published formulas. I do not see an independent error in the GPF derivation itself, and the authors state clear limitations near T_c and regarding anharmonic processes. Therefore the appropriate outcome is unchanged: the paper is CONDITIONAL, pending verification that the complex roots obtained from Eq. (19) correspond to physical poles of the retarded propagator.","tokens_in":17954,"tokens_out":10104,"duration_ms":98503,"concrete_test":"For the parameters of Fig. 3, compute the roots of det M_2b(q=0,z)=0 with Eq. (19) and, on the same parameter grid, evaluate the retarded spectral weight functions χ_pp^{(±)}(ω)=(1/π) Im Tr[Q_2b^{-1}(ω+i0^+)] over a fine frequency sweep across each window. A physical damped mode must appear as a Lorentzian peak in at least one such spectral function, with peak position matching Re z and half-width matching Γ/2; if no peak appears at the continued root, or the mismatch exceeds the linewidth, the continuation is selecting the wrong sheet. As an internal consistency control, repeat the procedure on the one-band pair-breaking mode of Phys. Rev. Lett. 122, 093403 and require exact agreement with the published damped branch; any algebraic error in implementing Eq. (19) would be immediately visible.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central new predictions—the BEC Leggett mode crossing the pair-breaking edge and acquiring damping, and the formation of a second damped root far from BEC—are obtained only from complex roots of Eq. (17), which exist solely after the analytic continuation defined in Eq. (19). Section III states explicitly that the uncontinued determinant has no complex roots; the roots are created by the branch-cut prescription f^{(R)}(z)=f(z)-2πiρ_f(z) for Im z<0. This is a sheet-selection assumption, not a proven result. The determinant has four non-analytic points (2Δ_j and 2√(Δ_j^2+μ^2)); the choice of which windows are connected to the physical retarded response fixes whether the 'damped' roots are resonances or artifacts of the continuation. The authors themselves acknowledge (Sec. III) that the second solution is 'strictly speaking' physically relevant only inside window B and that it is 'a complicated question' whether it is pair-breaking or Leggett in character. The spectral-weight contour plots in Sec. IV are qualitative and are not used to verify quantitatively that the complex roots coincide with peaks of the retarded spectral function. If Eq. (19) places roots on the wrong Riemann sheet, the finite-temperature damping predictions—the main advance over prior work—do not follow from the calculation.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript extends the Gaussian pair fluctuation (GPF) approach to a two-band Fermi superfluid at finite temperature and computes the Leggett collective mode spectrum in the long-wavelength limit. The authors write the two-band inverse GPF propagator explicitly (Eqs. (12)-(14)), solve the coupled gap equations (Eq. (8)), and use the Nozières analytic continuation of Eq. (19) to obtain complex poles of the fluctuation propagator, thereby determining both eigenfrequencies and damping factors. Their central results are two qualitatively different regimes: in the BEC regime, the Leggett mode frequency smoothly crosses the pair-breaking continuum edge and acquires a small damping, whereas far from the BEC regime the Leggett mode undergoes an avoided crossing with the pair-breaking edge and a second, strongly damped root appears. The mode assignment is supported by phase-phase and amplitude-amplitude spectral weight contours in Section IV, and the results are compared with the low-frequency effective field theory of Ref. [18]. The paper explicitly acknowledges that the mean-field equation of state is inaccurate near T_c and that anharmonic processes beyond GPF are omitted.","tokens_in":18204,"tokens_out":7336,"duration_ms":73617,"significance":"If the complex-pole identification is correct, this is a useful and timely advance: it provides finite-temperature frequencies and damping rates for Leggett modes beyond the low-energy expansion, which is necessary when the mode energy approaches the pair-breaking continuum. The calculation is transparent and internally consistent: the GPF action, the gap equations, and the analytic continuation are explicit, and the T=0 limit reproduces earlier results. The work also makes falsifiable predictions for cold-atom experiments near orbital Feshbach resonances and in the BCS-BEC crossover. The main risk is not circularity or fitted parameters but the physical interpretation of the analytic continuation: the paper itself states that the uncontinued determinant has no complex roots, so all damped-mode predictions rest on the sheet-selection prescription of Eq. (19).","major_comments":[{"comment":"The damped-mode results are load-bearing and depend entirely on the analytic continuation f^{(R)}(z)=f(z)-2πiρ_f(z) for Im z<0. The paper explicitly states that the uncontinued determinant has no complex roots; the roots are created by this branch-cut prescription. Since the determinant has four non-analytic points (2Δ_j and 2√(Δ_j²+μ_j²)), the continuation can select different Riemann sheets depending on which window is used, and the text later concedes that the second solution is strictly physically relevant only inside window B. The manuscript does not verify that the complex roots are poles of the physical retarded response rather than artifacts of this sheet choice. I request a quantitative check: compute the retarded spectral weight functions from Appendix A (for example χ_pp^(±) and χ_aa^(±)) on the real frequency axis for the same parameters as in Figs. 1-3 and show that the complex roots coincide with peaks in both frequency and width, at least inside each window. The contour plots in Section IV are qualitative and do not close this gap. Without such a check, the central predictions—the BEC Leggett mode crossing the pair-breaking edge and the second damped root far from BEC—are not fully established.","section":"III, Eqs. (17)-(19); Fig. 3"},{"comment":"The conclusion that in the BEC regime the Leggett frequency remains finite as T→T_c is obtained from mean-field background parameters. The authors acknowledge that the mean-field equation of state is not justified near T_c and argue that T/T_c scaling preserves crossings and anticrossings. That argument does not by itself fix the quantitative value of the frequency at T_c, which is a prominent result in the conclusions. I recommend testing robustness with a fluctuation-corrected equation of state in at least one representative BEC case, or explicitly stating that the finite value at T_c is a mean-field-level prediction whose quantitative magnitude could change when fluctuations are included.","section":"III, Eq. (20) and Fig. 2"},{"comment":"The second root ω_L^(B)-iΓ_L^(B)/2 is described as 'strictly speaking' physically relevant only inside window B, yet the text uses its formal continuation outside that window to conclude that 'the second root appears at lower temperatures starting from a finite ω_L^(B) with zero damping' and to build the avoided-crossing narrative. This extrapolates beyond the regime where the solution is claimed to be physical. Either restrict the claims to window B or provide evidence that the formal continuation corresponds to a resonance in the retarded response, as requested in the first major comment.","section":"III, paragraph on the second root; Fig. 3"}],"minor_comments":[{"comment":"The word 'reprersents' should be 'represents'.","section":"I, Introduction"},{"comment":"The phrase 'lower semi-panel' should read 'lower half-plane'.","section":"III, near Eq. (19)"},{"comment":"The definition \\tilde{Q}^{(2)} = \\tilde{Q}^{(2)} + κ Δ_1/Δ_2 I uses the same symbol on both sides; presumably the left-hand side is \\tilde{Q}^{(2)} built from Q^{(2)}, so the notation should be corrected to avoid an apparent self-referential definition.","section":"Appendix A, Eq. (A20)"},{"comment":"The contour plots would be easier to assess with an explicit color scale and clearly labeled peak positions, especially since the text refers to specific branches (Leggett, phononic, PB-1, PB-2) that are not all marked in every panel.","section":"IV, Figs. 4 and 5"}],"recommendation":"major_revision","confidential_remarks":"The main technical machinery is an extension of the authors' earlier one-band PRL method (Ref. [19]), but the two-band setting produces new physics, and the self-citation pattern is appropriate rather than excessive. The key risk is the physical-sheet selection for the complex poles; if the requested spectral-function verification succeeds, the paper should be publishable. I do not see any circularity or data-quality issue."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"I read the Klimin-Kurkjian-Tempere paper on Leggett modes at finite temperature. The genuinely new content is the nonperturbative calculation of complex eigenfrequencies for the two-band Leggett mode using the GPF propagator, going beyond their earlier low-energy EFT. The clean result is that in the BEC regime the Leggett mode frequency stays finite at Tc and, for large enough detuning, crosses the pair-breaking edge and acquires damping; far from BEC there is an avoided crossing and a second, strongly damped root appears. The T=0 limit matches earlier work, and the equations and parameter sets are explicit enough for an independent group to re-implement.\n\nThe paper earns credit for being careful. The gap equations, GPF matrix elements, and analytic continuation are written out, and the authors openly note the mean-field equation of state is poor near Tc and that anharmonic processes are beyond the present approximation. That honesty is welcome.\n\nThe soft spot is exactly what the stress-test note identifies. The uncontinued determinant in Eq. (17) has no complex roots; all the damping predictions come from the analytic continuation in Eq. (19). The paper gives the Nozieres rationale for that continuation, but it does not prove that the chosen lower-half-plane sheet is the physical one, and the spectral-weight contour plots in Sec. IV are qualitative. They are not used to check that the computed poles coincide with peaks of the retarded spectral function. The authors themselves say the second root is 'strictly speaking' relevant only inside window B and that its physical interpretation is a complicated question. That is honest, but it leaves the central new predictions resting on a sheet-selection assumption. A referee should ask for a direct quantitative comparison of pole positions with spectral-peak maxima, or a clearer argument for the sheet choice.\n\nOne smaller issue: the statement that all crossings and anticrossings survive a more precise equation of state is asserted without proof. It may be true, but a mean-field calculation doesn't by itself justify that invariance.\n\nThis is a paper for people working on multiband superfluids, cold-atom or otherwise. It deserves peer review: the formalism is sound, the new result is concrete, and the sheet question is addressable rather than a deal-breaker. I would send it to referees.","headline":"Finite-temperature two-band Leggett modes: the BEC survival prediction is plausible, but the analytic-continuation sheet selection needs a quantitative check before the damping results are taken to the bank.","tokens_in":18801,"tokens_out":4606,"would_cite":true,"duration_ms":37828,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"At finite temperature, a two-band Fermi superfluid's Leggett mode either crosses the pair-breaking edge and survives as a damped mode (in the BEC regime) or avoids crossing and produces a second damped root (far from BEC).","keywords":["Leggett mode","two-band Fermi superfluid","BCS-BEC crossover","Gaussian pair fluctuations","pair-breaking continuum edge","analytic continuation","orbital Feshbach resonance","finite temperature"],"falsifier":"Measure the pair spectral response of a two-band Fermi gas while tuning band detuning and temperature. In the BEC regime the paper predicts a Leggett peak whose frequency crosses the pair-breaking edge and remains visible as a damped mode; far from BEC it predicts an avoided crossing and a separate damped peak above the edge. Directly comparing the continued-pole predictions with the observed response peaks—or with a numerical response calculation that does not use the branch-cut continuation—would settle whether the predicted complex roots are physical.","tokens_in":17745,"feed_emoji":"⚛️","tokens_out":7242,"duration_ms":64750,"temperature":0.7,"pith_summary":"The paper asks what happens to the Leggett mode—the collective oscillation of the relative phase between two paired fermion bands—when a two-band Fermi superfluid is heated toward its superfluid transition. Using the Gaussian pair fluctuation approximation, the authors compute complex eigenfrequencies of the fluctuation propagator without assuming the mode energy is small. They find two qualitatively different finite-temperature fates. In the BEC regime, relevant to cold atoms near an orbital Feshbach resonance, the Leggett frequency remains finite at T_c and crosses the pair-breaking continuum edge, continuing as a weakly damped mode. Far from BEC it instead undergoes an avoided crossing with the pair-breaking edge, and a second strongly damped eigenfrequency appears at higher temperatures. These results give finite-temperature signatures that go beyond the low-energy effective-field-theory description.","feed_headline":"Leggett modes stay finite at T_c in BEC two-band superfluids","feed_subtitle":"In BEC gases the mode survives to T_c; away from BEC it bends away and a damped partner appears.","key_machinery":"The central object is the inverse Gaussian pair-fluctuation propagator for two bands, a 4×4 matrix in the space of the two bands and their particle–hole partners whose determinant vanishes at collective-mode poles. Because the determinant evaluated on the real axis has branch cuts but no complex roots, the authors use the standard continuation f^(R)(z)=f(z)−2πi ρ_f(z) for Im z<0, built from the spectral function on the real axis; the pair-breaking edges split the frequency plane into windows, and each pole solution is associated with one window. This analytic continuation is what converts the search for eigenfrequencies into a calculation that yields both frequency and damping factor self-consistently, beyond a perturbative imaginary part computed at a real frequency.","core_discovery":"Within the Gaussian pair fluctuation approximation, the paper establishes that the Leggett mode's interaction with the pair-breaking continuum depends on the chemical potentials of the two bands, not just on the Josephson coupling. In the BEC regime, where both chemical potentials are negative and pair-breaking collective branches are absent, the Leggett frequency grows with detuning and crosses the lower pair-breaking continuum edge essentially without feature; the pole acquires a damping factor but the mode does not vanish. In the unitarity/BCS-type regime, the Leggett frequency rises toward 2Δ2 with temperature but never crosses it: the approach to the edge is an avoided crossing with the pair-breaking branch, and at sufficiently high temperature a separate, damped solution appears in the window between the two pair-breaking edges. The same calculation also yields spectral weight functions showing that Leggett and phononic branches are predominantly phase fluctuations, while the pair-breaking branches are amplitude-like, and that Leggett modes carry a non-negligible amplitude component away from the BCS limit.","pith_inferences":["Inference: the finite-T_c survival of the Leggett mode predicts a specific experimental signature in the pair or density response of a two-band gas near an orbital Feshbach resonance: a peak that tunes continuously through the pair-breaking threshold as detuning is increased.","Inference: because pair-breaking branches become visible at q=0 through anticrossing with Leggett modes, a two-band gas could be used to observe amplitude-like excitations that are dark in one-band systems.","Inference: the Gaussian pair fluctuation damping omits three- and four-phonon anharmonic processes, so at low temperatures the predicted damping factors are plausibly lower bounds; including those processes would likely broaden the same peaks.","Inference: the same semianalytic continuation could be applied to a BCS-type multiband superconductor model to predict Leggett-mode frequencies and damping in materials similar to MgB2, something the paper notes as a possible reformulation."],"forward_implications":["In the BEC regime the Leggett mode should be observable near T_c: its frequency does not vanish at the transition, and above the pair-breaking edge it survives as a damped mode with a finite spectral weight.","Far from BEC, effective-field-theory predictions that cross the pair-breaking edge without structure are unreliable; the full solution shows an avoided crossing instead.","A two-band Fermi superfluid can host up to four collective branches—phononic, Leggett, and two pair-breaking branches—and the pair-breaking branches can appear at zero momentum only through their anticrossing with the Leggett mode.","The spectral weight functions show that 'phase' and 'amplitude' labels are only approximate away from the BCS limit, with Leggett modes containing an amplitude component and the branch above the continuum edge mixing with phononic character."],"supporting_citations":[{"why":"Provides the two-band Gaussian pair-fluctuation effective action, gap equations, and Josephson coupling definitions used for the calculation.","marker":"[18]"},{"why":"Supplies the nonperturbative analytic-continuation method through the branch cut, adapted here to two bands.","marker":"[19]"},{"why":"Gives the orbital-Feshbach-resonance parameters and zero-temperature collective-mode frequencies that the present T=0 results reproduce.","marker":"[14]"},{"why":"Provides the zero-temperature Leggett-mode calculation for the orbital Feshbach resonance used as the T=0 benchmark.","marker":"[13]"},{"why":"Shows how the same complex-pole method treats phononic collective excitations, the companion branch studied alongside Leggett modes.","marker":"[20]"},{"why":"Defines the general analytic continuation of functions with a branch cut on the real axis that underlies the damped-root calculation.","marker":"[21]"},{"why":"Defines the Leggett mode as the relative-phase collective excitation in two-band superconductors.","marker":"[1]"}],"fun_headline_variants":["Leggett mode persists to T_c in BEC regime","Avoided crossing shapes Leggett mode in two-band gas","Chemical potentials govern Leggett mode's pair-breaking edge","Two-band superfluids: Leggett mode survives past BEC","Damped partner appears for Leggett mode off BEC"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The calculation assumes that continuing the fluctuation propagator through the branch cut in the complex frequency plane is the correct way to identify damped collective modes; if the continuation lands on the wrong side of the cut, the frequencies and damping factors it produces could be mathematical artifacts rather than physical excitations.","fun_headline_variants_meta":{"raw":{"variants":["Leggett mode persists to T_c in BEC regime","Avoided crossing shapes Leggett mode in two-band gas","Chemical potentials govern Leggett mode's pair-breaking edge","Two-band superfluids: Leggett mode survives past BEC","Damped partner appears for Leggett mode off BEC"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000427,"raw_usage":{"total_tokens":2155,"prompt_tokens":884,"completion_tokens":1271,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":500,"completion_tokens_details":{"reasoning_tokens":1190}},"tokens_in":500,"tokens_out":1271,"duration_ms":11141,"temperature":1.0,"reasoning_tokens":1190,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T10:06:17.056153+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the pair spectral response of a two-band Fermi gas while tuning band detuning and temperature. In the BEC regime the paper predicts a Leggett peak whose frequency crosses the pair-breaking edge and remains visible as a damped mode; far from BEC it predicts an avoided crossing and a separate damped peak above the edge. Directly comparing the continued-pole predictions with the observed response peaks—or with a numerical response calculation that does not use the branch-cut continuation—would settle whether the predicted complex roots are physical.","supporting_citations":[{"cited_title":"Collective modes in a two-band superfluid of ultracold alkaline-earth-metal atoms close to an orbital Feshbach resonance","cited_arxiv_id":null,"evidence_quote":"Provides the two-band Gaussian pair-fluctuation effective action, gap equations, and Josephson coupling definitions used for the calculation."},{"cited_title":"Pair-breaking collective branch in BCS superconduc- tors and superfluid Fermi gases","cited_arxiv_id":null,"evidence_quote":"Supplies the nonperturbative analytic-continuation method through the branch cut, adapted here to two bands."},{"cited_title":"Strongly Interacting Gas of Two-Electron Fermions at an Orbital Feshbach Resonance","cited_arxiv_id":null,"evidence_quote":"Gives the orbital-Feshbach-resonance parameters and zero-temperature collective-mode frequencies that the present T=0 results reproduce."},{"cited_title":"The work [14] reprersents an alternative method exploiting the density- density response function","cited_arxiv_id":null,"evidence_quote":"Provides the zero-temperature Leggett-mode calculation for the orbital Feshbach resonance used as the T=0 benchmark."},{"cited_title":"Coexis- tence of giant Cooper pairs with a bosonic condensate and anomalous behavior of energy gaps in the BCS-BEC crossover of a two-band superfluid Fermi gas","cited_arxiv_id":null,"evidence_quote":"Shows how the same complex-pole method treats phononic collective excitations, the companion branch studied alongside Leggett modes."},{"cited_title":"Strongly interact- ing Sarma superfluid near orbital Feshbach resonances","cited_arxiv_id":null,"evidence_quote":"Defines the general analytic continuation of functions with a branch cut on the real axis that underlies the damped-root calculation."},{"cited_title":"777, µ/ ∆2|T =0,δ =0 ≈ 1","cited_arxiv_id":null,"evidence_quote":"Defines the Leggett mode as the relative-phase collective excitation in two-band superconductors."}],"review_version":1}