{"id":"1a1af516-e387-46e5-a11f-2f6a4871267b","arxiv_id":"1908.07541","paper_version":2,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"A combined lattice and analytic study shows that the susceptibility term in a modified Lee-Weinberg equation does not regulate late-time freeze-out in a stop-like dark matter model, leaving mass splitting as the only equilibrium regulator.","lead":"This paper calculates a susceptibility that controls how bound states of dark matter mediator particles affect the annihilation rate in the early universe. It finds that in a widely studied stop-like model, this effect is too small to fix the dark matter abundance, so mass splitting remains the main regulator.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central no-regularization conclusion hinges on Eq. (3.9), a ground-state-only order-of-magnitude estimate; Fig. 1 validates it only at M_kin=14Λ, so a fuller bound-state sum could still move T^3 p̂2 into the regulating regime before S̄3 dominates.","rationale":"The negative conclusion is plausible and has real supporting evidence: the lattice measurement in Fig. 1 is a genuine non-perturbative check, and the authors are honest about the order-of-magnitude status of the analytic estimate. I also note that S̄3 and p̂2 share the same Boltzmann factor e^{βΔE}, which shields the conclusion from some normalization errors: even a relatively large multiplicative error in p̂2 would not automatically reverse the inequality because S̄3 has an extra (Mα/T)^3 factor. However, the shield is not complete. The criterion in Eq. (6.4) is a threshold condition, and the timing with which T^3 p̂2 reaches 1e11 relative to the growth of S̄3 controls the final abundance. That timing is exponentially sensitive to the binding energy and to the multiplicity of contributing states, and the single lattice point at M_kin = 14Λ does not probe the small-coupling, large-M/T regime that governs the cosmological conclusion. Therefore the reader's conditional verdict is appropriate: the central claim should be accepted with a concrete robustness check of the p̂2 estimate, rather than treated as a settled quantitative finding.","tokens_in":10591,"tokens_out":20012,"duration_ms":666394,"concrete_test":"Evaluate Eq. (3.8) at a cosmological point, e.g. M = 10 TeV, z = 10^4 and α = 0.1 with the same running prescription as Sec. 6, including the full Coulomb bound-state sum (all n up to n_cut with degeneracy n^2), octet channels, and hyperfine-shifted states; feed the resulting p̂2 into the relic-density integration of Sec. 6. If T^3 p̂2 is within about two orders of magnitude of Eq. (3.9), or if the final yield Y remains below roughly 1e-13 even when p̂2 is artificially multiplied by 100, the central conclusion survives. If the full-spectrum p̂2 is orders of magnitude larger and moves Y up to the observed value 0.12, the concern lands.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's quantitative criterion for a working regulator is 8 p̂2 n = 8 T^3 p̂2 (s/T^3) Y ≫ 1 at Y ≈ 1e-13 (Eq. (6.4)), i.e. T^3 p̂2 ≳ 1e11. T^3 p̂2 is obtained from Eq. (3.8) by replacing the entire two-particle spectral sum with one Coulomb ground state (Eq. (3.9)): cθ,η → d_s^2, octet and hyperfine contributions are dropped, and the sum over bound states is truncated to a single binding energy ΔE = α^2 M/4. The result is exponentially sensitive to α, and the lattice test in Fig. 1 measures only one point, M_kin = 14Λ, with large coupling α ≳ 0.3, far from the cosmological α ≈ 0.1 regime. Moreover, the integration in Sec. 6 uses the same analytic p̂2, so the agreement in Fig. 1 does not independently verify the late-time exponential growth. The authors themselves list this caveat in Sec. 7: the analytic value of p̂2 is only an order-of-magnitude estimate. If the true spectrum supplies even a moderate multiplicity factor or a slightly deeper effective binding energy, T^3 p̂2 could cross the 1e11 threshold at smaller z than in Fig. 2, changing the balance against S̄3. This is the weakest point of the central claim.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the susceptibility coefficient p̂2 that enters the modified Lee-Weinberg equation (1.2) when chemical potentials are eliminated in favour of the number density. Section 2 derives the Saha-type relation (2.12) from a fugacity expansion truncated at two-particle states. Section 3 gives an order-of-magnitude estimate, Eq. (3.9), in which the two-particle spectral sum is replaced by a single Coulomb-like ground state with binding energy ΔE = α²M/4, while octet and excited states are neglected. Section 4 formulates p̂2 as a non-perturbative lattice observable, Eq. (4.7), and Section 5 presents a single NRQCD lattice measurement at M_kin = 14Λ, comparing it with the analytic estimate at both vacuum and thermal coupling choices. Section 6 applies the same analytic p̂2 to a Majorana singlet dark matter model with a strongly interacting stop-like mediator, integrating Eq. (1.2) down to late times. The principal conclusion is that the susceptibility-induced regularization is ineffective: the combination 8 p̂2 n remains far below unity in the regime where the abundance Y ≈ 10⁻¹³ is set, so the viable parameter space continues to be controlled by the mass splitting ΔM rather than by p̂2. The paper closes by noting that the analytic value of p̂2 is only an order-of-magnitude estimate and that a full perturbative calculation would be of three-loop order.","tokens_in":10985,"tokens_out":9848,"duration_ms":102150,"significance":"The paper addresses a genuinely open technical question in thermal dark matter computations: whether the Saha-type susceptibility, which is in principle present in Eq. (1.2), can regularize the late-time growth of Sommerfeld-enhanced annihilation. The authors provide a concrete non-perturbative lattice definition of the susceptibility, Eq. (5.1), and take the first step toward measuring it with NRQCD, which is a meaningful methodological contribution. They also formulate a general low-temperature ratio S̄3/(T³ p̂2) ≃ (Mα/T)³ that gives a plausible reason for the inefficiency of the p̂2 regularization, and they explicitly flag the limitations of their estimate in Section 7. If the central conclusion holds, it rules out a potentially attractive solution to the overclose problem of coannihilating dark matter and sharpens the role of the mass-splitting regulator. The main limitation is that the central claim relies on an order-of-magnitude analytic estimate that is exponentially sensitive to α and is validated by only one lattice point with statistical-only errors and up to ~50% systematic uncertainty; the general ratio argument mitigates but does not fully remove this concern.","major_comments":[{"comment":"The central no-regularization conclusion rests on the single-ground-state estimate (3.9). The authors correctly stress that this is an order-of-magnitude estimate and that it is exponentially sensitive to α, but the manuscript does not quantify how much T³ p̂2 could change if a more complete treatment of the two-particle spectral sum were adopted, including excited Coulomb states, octet degrees of freedom, or a modestly different effective binding energy. Since the integration in Section 6 uses this same estimate, the agreement shown in Fig. 1 does not independently validate the late-time comparison on which the conclusion is based. I request a robustness scan: for the target yield Y ≈ 10⁻¹³ and the criterion 8 p̂2 n ≫ 1 from Eq. (6.4), please determine how large T³ p̂2 would need to be at the relevant z for the p̂2 regularization to affect the final abundance, and then state whether any plausible modification of Eq. (3.9) — for example, a multiplicity factor of 10–100 or a shift in ΔE within the uncertainty of α — can achieve that value before S̄3 drives Y below the observed value.","section":"Sec. 3, Eq. (3.9); Sec. 6, Eq. (6.4)"},{"comment":"The lattice test is limited to a single rest mass M_kin = 14Λ, in a regime where α ≳ 0.3, far from the cosmological value α ≈ 0.1 used in Section 6. The errors quoted are statistical only, and the authors themselves note that systematic uncertainties could be as large as ~50%. Moreover, the two analytic curves labeled 'vacuum coupling' and 'thermal coupling' are scaled versions of the same estimate (3.9), so the comparison in Fig. 1 is a test of the temperature dependence of a single-parameter form rather than an independent probe of the absolute normalization of T³ p̂2. I ask the authors to state explicitly how the central claim would be affected if T³ p̂2 were larger than Eq. (3.9) by a factor of 10 or 100 at the z values of interest (z ≈ 10³–10⁴), and to separate, in Fig. 1 or its caption, the systematic uncertainty in the lattice point from the spread between the two coupling prescriptions.","section":"Sec. 5, Fig. 1"},{"comment":"The general argument that S̄3/(T³ p̂2) ≃ (Mα/T)³ ≫ 1 is presented as the reason why p̂2 cannot compensate for S̄3. This ratio is derived from the same ground-state approximations, Eqs. (6.2) and (6.3), that underlie the quantitative estimate, so it inherits the uncertainties listed above. If the true bound-state spectrum supplies a large multiplicity factor to p̂2 but a smaller one to S̄3, the ratio could be shifted substantially. Please clarify to what extent this argument is parametric (i.e., would remain valid for any spectral sum with the same leading exponential behaviour) and to what extent it depends on the single-ground-state approximation.","section":"Sec. 7"}],"minor_comments":[{"comment":"The word 'estime' in the final paragraph should be 'estimate'.","section":"Sec. 7"},{"comment":"The notation d²_s appears without explicit definition of the spin degeneracy factor in Eq. (3.8); please spell out that d_s ≡ 2s + 1 at the first use of c_{θ,η} → d²_s.","section":"Sec. 3, Eq. (3.9)"},{"comment":"The caption states that 'the errors of the lattice results are statistical only', but the figure does not show error bars. I recommend adding explicit error bars (or the systematic band) to the figure, since the quoted ~50% uncertainty is essential to the calibration claim.","section":"Sec. 5, Fig. 1"},{"comment":"The combination 8 p̂2 n is introduced abruptly; a sentence explaining that this is the quantity controlling the correction term in Eq. (2.12) would improve readability.","section":"Sec. 6, Eq. (6.4)"}],"recommendation":"major_revision","confidential_remarks":"The paper is interesting and likely correct in its qualitative conclusion, but the central claim is more sensitive to the approximate nature of Eq. (3.9) than the manuscript acknowledges. The requested robustness scan is feasible within the scope of a revision and would turn the conclusion from a plausible order-of-magnitude statement into a defended one. I do not see evidence of circularity or of the conclusion being presupposed by the lattice measurement; the concern is purely that the calibration is too weak to pin down the absolute size of T³ p̂2 in the cosmological regime."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this paper does something new—it measures a susceptibility p̂2 on the lattice and then, for the first time, includes it in a full freeze-out computation for a stop-like dark matter model. The main result is a negative one: p̂2 does not regularize the late-time abundance when the mass splitting is small (2ΔM < ΔE). That is a useful clarification for model builders, because prior work simply set p̂2=0 and leaned entirely on ΔM.\n\nThe lattice formulation, eq. (5.1), is a genuine step forward. The comparison in fig. 1 is not a precision test, but it does give non-perturbative support to the analytic estimate at one point, and the authors are honest about its status. The general argument at the end, \\bar S_3/(T^3 \\hat p_2) ~ (Mα/T)^3 ≫ 1, is actually stronger than the numerical estimate because it makes the negative conclusion less dependent on the exact value of p̂2.\n\nThe soft spots are real. The lattice data are one mass point, M_kin=14Λ, with large coupling α≳0.3, far from the cosmological α≈0.1, and only statistical errors are reported—the authors say systematic could be ~50%. The analytic p̂2 in eq. (3.9) keeps only a single Coulomb ground state and drops octet and hyperfine contributions. The stress-test note is right that a larger true p̂2 could cross the 8p̂2n≳1 threshold earlier and shift the balance against S̄3. That is the weakest point of the central claim. But the paper already flags it in sec. 7, and the (Mα/T)^3 ratio would have to be badly wrong for the conclusion to flip. I don't think the caveat is a deal-breaker; it is exactly the kind of uncertainty that should be stated and, ideally, quantified in a revised version.\n\nWho gets value from this: dark matter modelers working on strongly interacting coannihilation, and lattice people interested in susceptibilities. It deserves a serious referee. I would send it to review and ask the authors to add a sensitivity scan of the conclusion to p̂2 (how much would it need to grow to reverse the claim) and a clearer treatment of the lattice systematics. That is revision work, not grounds for rejection.","headline":"New lattice measurement of p̂2 and a plausible negative result for its role in freeze-out; the quantitative uncertainty is real but does not overturn the main conclusion.","tokens_in":11468,"tokens_out":11255,"would_cite":true,"duration_ms":143203,"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":"This paper argues that in a stop-like dark matter model, the susceptibility correction $\\hat p_2$ does not efficiently regularize the late-time freeze-out abundance when $2\\Delta M < \\Delta E$, so the viable domain remains controlled by…","keywords":["dark matter freeze-out","non-relativistic susceptibility","Saha equation","Sommerfeld enhancement","bound states","non-relativistic lattice QCD","Majorana singlet dark matter","coannihilation"],"falsifier":"Compute $T^3\\hat p_2$ at the couplings and values of $z = M/T$ relevant to cosmology, either through the full three-loop perturbative evaluation of eq. (4.7) or through lattice simulations at smaller coupling; if for $M \\sim 1\\text{--}20$ TeV and $z \\sim 10^3$ it exceeds $10^{11}$, then $8\\hat p_2 n$ can reach order unity at the yields of interest and the regularization would be effective, contrary to the paper's conclusion.","tokens_in":10402,"feed_emoji":"⚛️","tokens_out":11013,"duration_ms":90083,"temperature":0.7,"pith_summary":"The paper asks whether the non-relativistic susceptibility correction $\\hat p_2$, which implements Saha-type ionization equilibrium in the modified Lee-Weinberg equation, can regularize late-time dark matter freeze-out when bound states enhance annihilation. Studying a Majorana singlet dark matter particle co-annihilating with a strongly interacting stop-like scalar, the authors conclude that it cannot: whenever the mass splitting obeys $2\\Delta M < \\Delta E$, the thermally averaged Sommerfeld factor grows faster than $\\hat p_2$ and drives the abundance to very small values before $\\hat p_2$ becomes relevant. An order-of-magnitude estimate based on a single Coulomb-like bound state, with binding energy $\\Delta E = \\alpha^2 M/4$, is supported by a non-perturbative lattice measurement, giving confidence in the extrapolation to the cosmological regime. If this conclusion holds, the viable domain of such models remains controlled by the mass splitting $\\Delta M$, not by the susceptibility. The result also clarifies where the inclusive chemical-potential framework is reliable and where resolved bound-state dynamics are necessary.","feed_headline":"Susceptibility can't rescue multi-TeV dark matter freeze-out","feed_subtitle":"The Saha-type susceptibility leaves the mass splitting as the only equilibrium control in stop-like dark matter models.","key_machinery":"The central object is the susceptibility-normalized coefficient $\\hat p_2$, defined through $p_2 = \\hat p_2 n_{\\rm eq}^2 T$, which closes the modified Lee-Weinberg equation by expressing the chemical potential in terms of the density through $e^{\\beta\\mu} n_{\\rm eq} = 2n/(1+\\sqrt{1+8\\hat p_2 n})$. Analytically it is estimated as $T^3\\hat p_2 \\simeq (2/N_c^2)(\\pi T/M)^{3/2}(e^{\\beta\\Delta E}-1)$, assuming one Coulomb-like bound state of the stop-antistop system with binding energy $\\Delta E = \\alpha^2 M/4$. Non-perturbatively it is measured from the disconnected heavy-propagator correlator in eq. (5.1), using non-relativistic lattice QCD; the connected part cancels, and the gauge-field lines linking the two propagators are what carry the bound-state physics. The comparison between the two estimates, with vacuum-like coupling at low temperature and thermal coupling at high temperature, is what licenses using the analytic form in cosmology.","core_discovery":"The paper's central claim is that the coefficient $\\hat p_2$, defined through $p_2 = \\hat p_2 \\, n_{\\rm eq}^2 T$ and entering the number density through $e^{\\beta\\mu} n_{\\rm eq} = 2n/(1+\\sqrt{1+8\\hat p_2 n})$, does not provide an efficient regularization of the late-time abundance in the stop-like model. The mechanism is quantitative: for $\\hat p_2$ to matter one needs $8\\hat p_2 n \\gtrsim 1$, which for $Y \\simeq 10^{-13}$ requires $T^3\\hat p_2 \\gg 10^{11}$; such values are reached only at $z \\equiv M/T$ so large that the Sommerfeld factor $\\bar S_3$ has already exceeded $10^{10}$ and pulled the yield down. The authors establish this with three ingredients: an order-of-magnitude estimate of $\\hat p_2$ from a Coulomb-like ground state, a lattice measurement in large-coupling QCD that agrees with the estimate, and an explicit integration of the modified rate equation for $M = 1\\ldots 500$ TeV. The stated conclusion is that in practice the regularization by $\\hat p_2$ is insufficient to make the system viable if $2\\Delta M < \\Delta E$, leaving $\\Delta M$ as the only possible equilibrium regulator.","pith_inferences":["If additional bound states or excited states contribute to $\\hat p_2$ more than assumed, the estimate could be systematically low; a full three-loop calculation and lower-coupling lattice data would be the direct test.","The same ratio $\\bar S_3/(T^3\\hat p_2)\\sim (M\\alpha/T)^3$ suggests that in models with weaker Sommerfeld enhancement or stronger binding, $\\hat p_2$ could still provide the regularizing effect the authors find absent here.","The lattice method for measuring the disconnected correlator could be applied to other non-relativistic dark sectors, giving a non-perturbative route to chemical equilibration rates that does not require enumerating bound states."],"forward_implications":["If the paper is right, models with strongly interacting mediators must satisfy $2\\Delta M > \\Delta E$ to avoid unacceptably large late-time annihilation, since $\\hat p_2$ cannot do the job.","The viable dark matter mass range extends at least to the multi-TeV domain, as previously found, but only when such a mass splitting is present.","At late times the ratio $\\bar S_3/(T^3\\hat p_2) \\simeq (M\\alpha/T)^3 \\gg 1$, so the suppression mechanism is generic rather than accidental in this model.","The lattice test at large coupling supports using the analytic $\\hat p_2$ estimate at small coupling, meaning the negative conclusion is not an artifact of the extrapolation, within the uncertainty of the estimate."],"supporting_citations":[{"why":"introduces the modified Lee-Weinberg equation with chemical potential and the Saha-type expression that $\\hat p_2$ parametrizes","marker":"[10]"},{"why":"supplies the non-relativistic lattice framework and propagator construction used for the measurement","marker":"[9]"},{"why":"provides the stop co-annihilation model, the thermal mass difference, and the Sommerfeld factors used in the cosmology application","marker":"[25]"},{"why":"gives the lattice setup and prior measurements in the same temperature range that support the comparison","marker":"[13]"},{"why":"fixes the low-temperature Coulombic coupling scale used in the binding-energy estimate","marker":"[18]"},{"why":"defines the phenomenologically allowed model region and the role of the mass splitting","marker":"[27]"}],"fun_headline_variants":["Susceptibility fails to save multi-TeV dark matter","New susceptibility study: no help for multi-TeV freeze-out","Stop-like dark matter stuck without mass splitting","Multi-TeV dark matter needs tight mass splitting","Susceptibility won't fix late-time annihilation"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that a single Coulomb-like ground state with binding energy $\\Delta E = \\alpha^2 M/4$ dominates the susceptibility, making the estimate exponentially sensitive to the coupling; the lattice test is performed at large coupling and moderate $M/T$, far from the small-coupling cosmological regime, so a much larger true $\\hat p_2$ could overturn the conclusion.","fun_headline_variants_meta":{"raw":{"variants":["Susceptibility fails to save multi-TeV dark matter","New susceptibility study: no help for multi-TeV freeze-out","Stop-like dark matter stuck without mass splitting","Multi-TeV dark matter needs tight mass splitting","Susceptibility won't fix late-time annihilation"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000164,"raw_usage":{"total_tokens":1265,"prompt_tokens":981,"completion_tokens":284,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":597,"completion_tokens_details":{"reasoning_tokens":209}},"tokens_in":597,"tokens_out":284,"duration_ms":3959,"temperature":1.0,"reasoning_tokens":209,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T12:04:31.312953+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute $T^3\\hat p_2$ at the couplings and values of $z = M/T$ relevant to cosmology, either through the full three-loop perturbative evaluation of eq. (4.7) or through lattice simulations at smaller coupling; if for $M \\sim 1\\text{--}20$ TeV and $z \\sim 10^3$ it exceeds $10^{11}$, then $8\\hat p_2 n$ can reach order unity at the yields of interest and the regularization would be effective, contrary to the paper's conclusion.","supporting_citations":[{"cited_title":"Rapid thermal co-annihilation through bound states in QCD","cited_arxiv_id":"1602.08105","evidence_quote":"supplies the non-relativistic lattice framework and propagator construction used for the measurement"},{"cited_title":"Studies of a thermally averaged p-wave Sommerfeld factor","cited_arxiv_id":"1904.07882","evidence_quote":"gives the lattice setup and prior measurements in the same temperature range that support the comparison"}],"review_version":1}