{"id":"5376d378-d990-435b-92e9-1966f7a9fa70","arxiv_id":"2501.09647","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"A full four-body final-state treatment of secluded and catalyzed dark matter annihilation weakens Fermi-LAT and Planck limits, reopening parameter space for scalar, fermion, and vector dark matter in two portal models.","lead":"This paper recalculates Fermi-LAT and Planck constraints on dark matter models that annihilate through a light mediator, using the full subsequent decay into standard model particles. It finds these constraints are weaker than previously thought, leaving more room for such dark matter models.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The relic-density curves that set the quoted mass bounds assume T_DM = T_SM, but the Appendix B scalar that enforces this has unspecified mass, decay width, and cosmological effects; the bounds are conditional on that patch.","rationale":"The paper's central claim is that a full 2DM -> 2A' -> 4SM calculation weakens the Fermi-LAT and CMB constraints compared with the simplified 2DM -> 2SM treatment, and the headline numbers are the resulting lower mass bounds in the catalyzed and secluded scenarios. Those bounds are obtained by combining the g_D vs. m_DM relic-density band (green in Fig. 11) with the independently derived exclusion curves. The Fermi-LAT and CMB pipelines are validated against official results, so the exclusion curves themselves are credible. The softest link is the relic-density band: it assumes T_DM = T_SM throughout freeze-out, which is not guaranteed by the tiny kinetic mixing. The authors recognize this and propose the Psi scalar in Appendix B, but the proposal is underspecified: without a mass, a decay width, and a check of its cosmological consequences, the patch is not a model but a placeholder. This is load-bearing because the quantitative conclusions (e.g., m_Phi > 709 GeV) would shift if the relic-density curve moved. The concern is not that the paper is wrong in a way that invalidates the direction of the effect; rather, the specific numerical claims are conditional on an unvalidated assumption. The reader identified this same weakest assumption, and the appropriate verdict remains CONDITIONAL: the physics insight is plausible and well-supported in the exclusion-curve part, but the appendix patch needs to be completed before the quoted limits can be taken at face value.","tokens_in":20865,"tokens_out":3909,"duration_ms":45105,"concrete_test":"Specify a complete model for Psi, including its mass m_Psi and an explicit decay operator (e.g., a small mixing with the SM Higgs or a tiny kinetic mixing to a SM gauge boson that gives Gamma_Psi), and recompute the thermal history including Psi. Verify that (i) the elastic-scattering condition (B3) is satisfied with lambda_PhiPsi ~ 10^-3, (ii) the 2Phi -> 2Psi cross section remains below 10% of the 2Phi -> 2A' rate (eta < 0.1), (iii) Psi decays before BBN (Gamma_Psi >> H(T ~ MeV)) and its decay products do not violate the Fermi-LAT and Planck constraints used in the paper, and (iv) the DM yield from Eqs. (2)-(3) changes by less than 10% relative to the T_DM = T_SM assumption. If any condition fails, the relic-density curve and the resulting mass bounds (709 GeV, 16 GeV, etc.) must be recomputed.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative results (e.g., for complex scalar DM with r=1.2, m_Phi > 709 GeV in the catalyzed case and > 16 GeV in the secluded case) are read off the intersection of the relic-density curve with the Fermi-LAT and CMB exclusion contours. Any shift in the relic-density calculation therefore moves the quoted mass bounds. The relic density is computed from the Boltzmann equations (2)-(3) under the assumption that the dark sector maintains kinetic equilibrium with the SM bath (T_DM = T_SM). The authors explicitly note in Section II.A (p.7) that this is difficult for s_epsilon as small as 10^-10, and Appendix B patches the issue by introducing a scalar Psi coupled to DM through the quartic term (B1). However, the patch is incomplete: the mass of Psi is not specified, no decay operator is written down (the quartic (B1) does not make Psi unstable), and the abundance, decay width, entropy injection, and BBN/CMB/Neff effects of Psi are not analyzed. Without a specified decay width, the requirement that Psi 'decay into SM particles after freeze-out' is vacuous: if Psi is long-lived or too abundant, it can alter the expansion history, inject entropy into the dark sector, or contribute to the DM density, changing the relic-density curve and hence the headline mass limits. Until a complete, consistent Psi sector is provided and shown not to disturb the thermal history, the claimed relaxation of constraints to specific masses is conditional.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript studies thermal relic dark matter in a U(1)_D dark photon model with complex scalar dark matter, and with fermionic and vector extensions, focusing on secluded annihilation (2DM -> 2A' -> 4SM) and catalyzed annihilation (with 3A' -> 2DM). The relic density is computed from coupled Boltzmann equations with the gauge coupling g_D tuned to reproduce Omega h^2 = 0.12, and gamma-ray energy spectra are generated with Pythia8 for the full 2A' -> 4SM chains in the U(1)_D x U(1)_Y and U(1)_D x U(1)_{L_mu-L_tau} portal models. Using Fermi-LAT 14.3-year 42-dSph likelihoods and the Planck 2018 p_ann bound, the authors derive upper limits on <sigma v> and g_D, finding considerably weaker constraints than the simplified 2DM -> 2SM analyses. For complex scalar DM with r = 1.2, the catalyzed lower mass bound becomes 709 GeV (1052 GeV) and the secluded bound 16 GeV (59 GeV) for the L_mu-L_tau (hypercharge) portal.","tokens_in":21272,"tokens_out":8181,"duration_ms":95922,"significance":"The main contribution is a more realistic treatment of indirect-detection constraints for secluded and catalyzed dark matter: replacing single-channel 2DM -> 2SM limits with full 2DM -> 2A' -> 4SM spectra changes the limits by factors of a few to several. The analysis is carefully validated by reproducing the official Fermi-LAT b bbar and tau+ tau- limits and the Planck 2015 bounds from Ref. [58], and the use of Pythia8 with electroweak showers is appropriate for this problem. The comparison of the hypercharge and L_mu-L_tau portals is physically well motivated, and the conclusion that the leptophilic portal is less constrained is robust in direction. If the relic-density assumptions are completed, the paper gives falsifiable mass and coupling targets for Fermi-LAT, CTA, and Planck. The central weakness is that the mass limits quoted in Section IV depend on a kinetic-equilibrium patch that is not fully specified, as detailed below.","major_comments":[{"comment":"The assumption that the dark sector remains in kinetic equilibrium with the SM bath until freeze-out, so that T_DM = T_SM, is load-bearing for the relic-density curves in Fig. 5 and Fig. 11 and hence for the quoted mass bounds such as 709 GeV and 16 GeV. Appendix B attempts to justify this with the scalar Psi and the quartic interaction (B1), but the patch is incomplete: the mass of Psi is not specified, no decay operator is written down, and the quartic (B1) does not by itself make Psi unstable. The abundance of Psi, its decay width, the decay temperature, and the resulting entropy injection or contribution to the dark matter density are not analyzed. If Psi is long-lived or over-abundant, it can alter the expansion history and shift the relic-density curve, moving the headline mass limits. Please either provide a complete Psi sector with mass and decay operators and show that BBN, CMB, and Delta N_eff constraints are satisfied, or demonstrate that the quoted limits are insensitive to relaxing the T_DM = T_SM assumption.","section":"Section II.A and Appendix B"},{"comment":"In the catalyzed regime with s_epsilon as small as 10^-10, the A' lifetime can be much longer than the DM freeze-out time. The Boltzmann system (2)-(3) tracks the A' number density but does not include the effect of A' decay products on the SM bath temperature or the dilution of the DM yield by late entropy injection. If the A' abundance at freeze-out is non-negligible, the subsequent decays inject entropy and reduce the final DM relic density, so the values of g_D tuned to Omega h^2 = 0.12 in Fig. 5(a) and Fig. 11(a) would change. Please quantify the A' yield at freeze-out, the A' decay temperature, and the resulting dilution for the catalyzed benchmarks, or restrict the claimed mass limits to the parameter region where this effect is negligible.","section":"Section II.A, Eqs. (2)-(3), and Fig. 5(a)"},{"comment":"The Fermi-LAT and Planck upper limits on <sigma v> are presented in Fig. 10 as functions of m_DM only, but in the 2DM -> 2A' -> 4SM chain the final-state spectra depend on the mediator mass, i.e. on r = m_DM/m_A', through the boost of A' and the energies of its decay products. The text does not state the value of r used to generate the spectra in Fig. 6 and the limits in Fig. 10, and Fig. 11(b) applies constraints across a range of r without describing any recomputation of the spectra. If the same <sigma v> limits are used for all r, the r-dependence of the gamma-ray and CMB deposition spectra is neglected, which would affect the exclusion curves. Please state the r value assumed in Fig. 10 and specify how the limits are recomputed for each r shown in Fig. 11(b).","section":"Section III.A and Fig. 10"}],"minor_comments":[{"comment":"The direct-detection bound in Eq. (8) is derived assuming m_A' approximately equal to m_Phi, but the benchmarks used throughout the paper have r = m_Phi/m_A' = 1.2; please use m_A' = m_Phi/r so that the numerical coefficient is consistent with the parameter space studied.","section":"Section II.A, Eq. (8)"},{"comment":"The captions of Figs. 6 and 10 should state the mass ratio r used in the spectral simulations; without this information the reader cannot tell whether the plotted constraints are meant to be universal or benchmark-specific.","section":"Section III.A, Figs. 6 and 10"},{"comment":"The text refers to 'PPPC4DM' while the cited code is 'PPPC 4 DM ID'; please use the official name consistently.","section":"Section III.A"},{"comment":"The relation between the horizontal axis Gamma_A' and the model parameters s_epsilon or s'_epsilon is not stated in Fig. 11(c); since the two portal models have different decay widths for the same mixing angle, please specify which relation is used or state that Gamma_A' is treated as a free parameter.","section":"Section IV, Fig. 11(c)"}],"recommendation":"major_revision","confidential_remarks":"The technical core of the indirect-detection analysis is sound and the comparison with simplified 2DM -> 2SM limits is a useful contribution. My main concern for the editor is that the headline mass limits are quoted as robust numbers although the relic-density calculation relies on an incomplete dark-sector thermal model in Appendix B and an unquantified A' decay/entropy-injection history. These issues are fixable with additional calculation, but they affect the central quantitative claims, so I cannot support acceptance in the current form. The ambiguity about the mass ratio used for the spectral constraints should also be resolved before publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The short version: this paper recalculates Fermi-LAT and Planck constraints for secluded and catalyzed dark matter using the complete 2DM -> 2A' -> 4SM annihilation chain rather than the simplified 2DM -> 2SM channels. The main result is that the old single-channel limits were too strong, roughly by factors of two to six in DM mass, and the U(1)_D x U(1)_{L_mu-L_tau} portal ends up with the weakest constraints. I think the qualitative conclusion is right and the calculation is worth taking seriously.\n\nWhat is genuinely new is the full four-body treatment with mixed hadronic/leptonic final states and electroweak shower corrections for these specific portals. The authors use Pythia8 to generate the gamma-ray spectra, which is appropriate, and they validate their pipelines by reproducing the official Fermi-LAT and Planck limits for b bbar and tau+ tau-. The cross sections come from the existing peer-reviewed literature, and the relic density follows the standard coupled Boltzmann equations. They are also honest about conventions and about the fact that the model-building ingredients are not new.\n\nThe soft spot is the one the reader flagged: kinetic equilibrium. With kinetic mixing s_epsilon as small as 10^{-10}, the dark sector would decouple from the SM bath long before freeze-out, and the Boltzmann equations assume T_DM = T_SM. The patch in Appendix B, a new scalar Psi with a quartic coupling to DM, is underspecified: no mass, no decay operator, no cosmological analysis. The quartic term (B1) does not make Psi unstable, and a long-lived or over-abundant Psi could alter the expansion history or contribute to the DM density. So the headline mass limits, like m_DM > 709 GeV for the catalyzed complex scalar case, are conditional on that patch. That said, the old analyses they compare against make the same assumption, so the relative relaxation of constraints is on firmer ground than the absolute numbers. The authors should either complete the Psi sector or explicitly frame the results as contingent on kinetic equilibrium.\n\nOverall: a solid phenomenological paper, not a paradigm shift. It deserves a serious referee. I would send it to review, and I would ask the referee to focus on Appendix B and on how sensitive the relic-density curves are to the dark-to-SM temperature ratio.","headline":"Full 4-body treatment relaxes Fermi-LAT/Planck limits for secluded and catalyzed DM, but the quoted mass bounds rest on an unfinished kinetic-equilibrium patch.","tokens_in":21757,"tokens_out":3676,"would_cite":true,"duration_ms":40749,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["95.35.+d","98.70.Rz","98.80.Cq"],"model":"deepseek-v4-flash","headline":"The paper claims that the correct annihilation chain for secluded dark matter is 2DM → 2A' → 4SM, and that using this full chain weakens Fermi-LAT and Planck bounds enough to reopen DM masses previously excluded.","keywords":["secluded dark matter","catalyzed annihilation","dark photon mediator","kinetic mixing","Fermi-LAT gamma rays","CMB constraints","complex scalar dark matter","U(1) L_mu-L_tau portal"],"falsifier":"A concrete check would be to include the thermal evolution of the Ψ scalar: if a population of light Ψ particles is needed to maintain kinetic equilibrium, its contribution to ΔN_eff or its decay products would distort the Planck CMB spectra, and the quoted mass windows would close. Alternatively, recomputing the Fermi-LAT limits with the mediator's finite decay length, so that some A' decays occur outside the dwarf galaxies, would change the J-factor weighting and could either tighten or further relax the bounds.","tokens_in":20684,"feed_emoji":"🔭","tokens_out":6492,"duration_ms":63326,"temperature":0.7,"pith_summary":"The paper argues that previous Fermi-LAT and CMB constraints on secluded and catalyzed dark matter used the wrong annihilation final state: they assumed 2DM → 2SM, whereas in these models the mediator A' is nearly as heavy as the dark matter and the actual chain is 2DM → 2A' → 4SM. Computing the gamma-ray yield and ionizing energy injection for the full chain, the authors find the bounds are weaker, and therefore much of the parameter space previously excluded is viable again. For a complex scalar dark matter with mediator mass ratio r = 1.2, masses above 709 GeV (catalyzed) and 16 GeV (secluded) survive all constraints in the U(1)$_D$ × U(1)$_{L_\\mu-L_\\tau}$ model, while the U(1)$_D$ × U(1)$_Y$ model is more restrictive. The result matters because it shows how model-dependent indirect limits are when mediators decay into mixed hadronic and leptonic final states.","feed_headline":"Full 4-particle annihilation weakens dark matter limits","feed_subtitle":"Counting the real 2DM→2A'→4SM chain reopens masses Fermi-LAT and Planck had excluded.","key_machinery":"The central object is the annihilation chain 2DM → 2A' → 4SM together with the mediator decay width $\\Gamma_{A'}$, which interpolates between the secluded regime (prompt decay) and the catalyzed regime (long-lived mediator with 3A' → 2DM). The paper computes thermally averaged cross sections $\\langle\\sigma_2 v\\rangle$ and $\\langle\\sigma_3 v^2\\rangle$ for scalar, Dirac, and vector dark matter, solves the coupled Boltzmann equations for the DM and A' number densities, and feeds Pythia8 spectra into the Fermi-LAT 42-dSphs joint likelihood and the Planck $f_\\text{eff}$-based $p_\\text{ann}$ bound. The key identity enabling the weakening is that the mediator is almost degenerate with the DM (r ≈ 1–1.5), so the four-body final states contain neutrinos and mixed leptonic and hadronic products that radiate fewer gamma rays per annihilation than the pure channels used previously.","core_discovery":"In the U(1)$_D$ dark-photon framework the dominant DM annihilation is 2Φ → 2A' followed by A' → SM, i.e. 2DM → 2A' → 4SM, with a 3A' → 2Φ process also active when the A' is long-lived (catalyzed annihilation). By generating the complete gamma-ray spectra of the four-body final states with Pythia8, including electroweak showers, and computing the CMB deposition efficiency $f_\\text{eff}$ for the same spectra, the paper derives 95% CL upper limits on $\\langle\\sigma v\\rangle$ that are uniformly weaker than the simplified single-channel $b\\bar{b}$ or $\\tau^+\\tau^-$ limits. In the leptophilic U(1)$_D$ × U(1)$_{L_\\mu-L_\\tau}$ model the mediator decays only to μ, τ, and neutrinos, so the constraints are the weakest. Consequently, with r = 1.2 the catalyzed complex-scalar scenario survives for $m_\\Phi \\gtrsim 709$ GeV in that model and $m_\\Phi \\gtrsim 1052$ GeV in the U(1)$_D$ × U(1)$_Y$ model; the secluded scenario survives for $m_\\Phi \\gtrsim 16$ GeV and 59 GeV respectively. Equivalent relaxations hold for Dirac fermion DM (catalyzed lower limits about 665 versus 910 GeV) and for vector DM (the catalyzed Fermi-LAT limit is relaxed from about 4.4 TeV to 706 GeV).","pith_inferences":["The same full-chain treatment applied to other gauge extensions such as U(1)$_{L_e-L_\\mu}$, U(1)$_{B-L}$, or other leptophilic portals would map out a spectrum of model-dependent indirect limits, with the ordering (leptonic weakest) robust while the absolute masses shift.","Because the mediator is nearly degenerate with dark matter, the off-shell and t-channel contributions in 2DM → 2A' are kinematically special; future gamma-ray observatories sensitive to the 10 GeV-to-TeV range could probe the r > 1 region that this paper finds open.","The auxiliary scalar Ψ required to maintain kinetic equilibrium is a testable input: its coupling $\\lambda_{\\Phi\\Psi}\\sim 10^{-3}$ and mass must satisfy BBN and CMB bounds, but its mass, decay width, and cosmological effects are left unanalyzed here.","The paper assumes prompt A' decay inside dwarf galaxies; a long-lived mediator that decays outside the dwarf would change the J-factor weighting and could either tighten or further relax the gamma-ray limits depending on the decay length."],"forward_implications":["The simplified single-channel bounds (b\\bar{b} or τ+τ−) that previously excluded secluded and catalyzed DM should not be used; the full four-body chain is required.","Leptophilic portals such as U(1)$_{L_\\mu-L_\\tau}$ are systematically less constrained than hadrophilic portals, so surviving DM candidates favor mediators that decay to muons, taus, and neutrinos.","Fermionic DM near 1 TeV in the catalyzed annihilation scenario, excluded in prior work, remains viable under the U(1)$_D$ × U(1)$_{L_\\mu-L_\\tau}$ model.","For vector DM with r = 1.2, the catalyzed Fermi-LAT limit is relaxed from about 4.4 TeV to 706 GeV.","The semi-catalyzed regime interpolates between the two extremes and has intermediate DM mass limits set by the value of $\\Gamma_{A'}$."],"supporting_citations":[{"why":"Supplies the non-relativistic cross-section formulas (E4) and (E5) used for the scalar and fermion DM cases, and the kinetic-equilibrium template adopted in Appendix B.","marker":"[15]"},{"why":"Introduced catalyzed annihilation for fermionic DM and the simplified 2DM-to-2SM limits that this work corrects.","marker":"[20]"},{"why":"Provides the vector DM catalyzed model, the cross sections used here, and the previous Fermi-LAT limit that is relaxed.","marker":"[21]"},{"why":"Gives the Planck 2018 cosmological parameters and the p_ann upper bound used for the CMB constraints.","marker":"[34]"},{"why":"Provides the 14.3-year Fermi-LAT legacy analysis of 42 dwarf spheroidals, including the J-factors and likelihood functions used to derive the gamma-ray limits.","marker":"[44]"},{"why":"Supplies the reference gamma-ray spectra for pure channels and the flux formula against which the Pythia8 spectra are compared.","marker":"[35]"},{"why":"The Pythia8 event generator is the central machinery used to produce the full 4SM gamma-ray spectra with electroweak showers.","marker":"[36]"},{"why":"Provides the feff/weighted-deposition method for CMB constraints that the paper reproduces and then updates to Planck 2018 data.","marker":"[58]"}],"fun_headline_variants":["4-body annihilation eases dark matter mass limits","Dark matter limits relax with full annihilation chain","Leptophilic dark matter dodges Fermi-LAT limits","Catalyzed annihilation reopens dark matter masses","Wider dark matter masses survive full annihilation calculation"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The analysis assumes the dark sector stays in kinetic equilibrium with the ordinary matter bath until freeze-out (T_DM = T_SM), even though the mixing angle is as small as $10^{-10}$, and this equilibrium is enforced by an auxiliary scalar Ψ introduced in Appendix B whose mass, decay width, and cosmological effects are not worked out.","fun_headline_variants_meta":{"raw":{"variants":["4-body annihilation eases dark matter mass limits","Dark matter limits relax with full annihilation chain","Leptophilic dark matter dodges Fermi-LAT limits","Catalyzed annihilation reopens dark matter masses","Wider dark matter masses survive full annihilation calculation"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000242,"raw_usage":{"total_tokens":1688,"prompt_tokens":1273,"completion_tokens":415,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":889,"completion_tokens_details":{"reasoning_tokens":344}},"tokens_in":889,"tokens_out":415,"duration_ms":4292,"temperature":1.0,"reasoning_tokens":344,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T19:47:56.570428+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A concrete check would be to include the thermal evolution of the Ψ scalar: if a population of light Ψ particles is needed to maintain kinetic equilibrium, its contribution to ΔN_eff or its decay products would distort the Planck CMB spectra, and the quoted mass windows would close. Alternatively, recomputing the Fermi-LAT limits with the mediator's finite decay length, so that some A' decays occur outside the dwarf galaxies, would change the J-factor weighting and could either tighten or further relax the bounds.","supporting_citations":[{"cited_title":"A Brief Review on Dark Matter Annihilation Explanation for $e^\\pm$ Excesses in Cosmic Ray","cited_arxiv_id":"0908.2908","evidence_quote":"Provides the 14.3-year Fermi-LAT legacy analysis of 42 dwarf spheroidals, including the J-factors and likelihood functions used to derive the gamma-ray limits."},{"cited_title":"Dark matter and spin-1 milli-charged particles","cited_arxiv_id":"1507.00571","evidence_quote":"Supplies the reference gamma-ray spectra for pure channels and the flux formula against which the Pythia8 spectra are compared."}],"review_version":1}