{"id":"d2594d68-fb2e-4f0d-9f85-274d478cb674","arxiv_id":"2411.15073","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Dark flavor symmetry stabilizes a subset of dark pions, whose velocity-suppressed co-annihilation allows them to be all of dark matter at GeV masses while evading indirect detection.","lead":"A composite dark matter model with a confining dark sector and a t-channel mediator is presented, where dark flavor symmetry makes some dark pions stable and the rest long-lived. The model opens up a viable GeV-mass dark matter window and predicts mixed semi-visible and emerging jet signatures at the LHC.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The exact dark-quark mass degeneracy m_Q = m δ (Sec. 2) is radiatively unstable because the portal breaks SU(4)_V; the paper asserts but never computes the induced πDM-πtran mass splitting, which must remain small for the velocity-suppressed cross section in Eq. (3.1) to control the 1-10 GeV window.","rationale":"Reading in good faith: the paper is a phenomenological study of a composite DM model, and its central claim is that n_f≥4 flavor symmetry stabilizes a subset of dark pions, that degenerate masses make annihilation velocity-suppressed, and that the resulting GeV-mass window is testable with dark showers. The strongest parts are the group-theoretic stability argument (App. A and B) and the concrete recast procedure with public code. The weakest load-bearing point is not the existence of dark baryons (acknowledged and plausibly subdominant by standard estimates) nor hadronization modeling (for the inclusive jet/MET searches used here, that uncertainty is plausibly mild); it is the unsupported smallness of the radiative mass splitting. If Δ is not small, Eq. (3.1) loses its p-wave form, the indirect-detection and CMB bounds strengthen, and the 'sneaky' 1-10 GeV window closes. The reader's weakest_assumption identifies the same place (degenerate mass matrix), so I partially agree; I sharpen it by noting the splitting is not merely hypothetical but radiatively induced by the same portal that makes the model testable. A one-loop calculation is a well-defined, bounded task and should be a condition for acceptance. The reader's CONDITIONAL verdict is therefore appropriate, and I do not change it.","tokens_in":27523,"tokens_out":18321,"duration_ms":207882,"concrete_test":"Compute the one-loop effective potential for the dark pions from the portal operator (2.5) and the κX terms in (2.2), and extract Δ^2 = m^2_πDM - m^2_πtran for representative n_f=4 benchmarks with κ=1, m_X=2 TeV, f_D=m_π, and m_π = 1 and 10 GeV. If |Δ|/m_π ≥ 0.1 for any benchmark used in Figs. 6-7, then Eq. (3.1) is not valid there, and the relic-density and indirect-detection curves in Fig. 1 must be regenerated including the mass splitting. If |Δ|/m_π remains below roughly 0.01, the concern is settled and the claimed GeV window survives.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central mechanism is the velocity-suppressed co-annihilation 2πDM→2πtran, whose p-wave form in Eq. (3.1) relies on all dark pions being exactly degenerate (Sec. 3.1). That degeneracy is guaranteed only by the tree-level choice m_Q = m δ_αβ, flagged as crucial in Sec. 2. The portal interaction (2.2) explicitly breaks SU(4)_V to SU(n_f-3)×U(1) (after SVD, only Q1-Q3 couple), so radiative corrections involving κ, X and SM quarks will split the dark quark masses and hence the pion masses. The paper acknowledges a loop-level splitting in Sec. 3.1 ('a small mass splitting will be generated ... not change the results provided Δ/mπD ≪ 1, which is expected'), but no calculation, bound, or estimate is given. This is not merely a UV tuning question: for the benchmark values used in the collider limits (κ=1, m_X=2 TeV), the splitting could be of order κ^2 m_X/(16π^2) or larger, which would invalidate Eq. (3.1) and the indirect-detection/CMB exclusions in Fig. 1 in the 1-10 GeV window. The stability of the DM pions themselves is protected by exact Q4 number, but their cosmological abundance and indirect detection signatures are not.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents a composite dark matter model consisting of an SU(N_d) confining dark sector with n_f dark quarks and a heavy scalar mediator X that couples the dark sector to SM quarks. For n_f >= 4, an unbroken dark flavor subgroup stabilizes a subset of dark pions, and for the n_f = 4 case (six stable pions pi_DM and nine transient pions pi_tran) the relic abundance is set by co-annihilation 2 pi_DM -> 2 pi_tran. Because the dark pions are degenerate, this cross section is claimed to be velocity suppressed, evading indirect detection and CMB bounds and opening a 1-10 GeV DM mass window. The paper derives the relic abundance, computes direct detection, indirect detection, CMB, and flavor constraints, and recasts LHC searches for four jets, jets plus missing energy, semi-visible jets, and emerging jets to set limits on the mediator mass.","tokens_in":27957,"tokens_out":9935,"duration_ms":98975,"significance":"If the central mechanism is sound, the model is an attractive minimal composite-DM benchmark: DM stability follows automatically from the flavor symmetry for n_f >= 4, the relic abundance depends mainly on the dark strong-sector parameters, and the p-wave suppression is a natural way to make GeV-scale thermal DM viable. The paper also has clear strengths: a group-theoretic proof of the accidental Z_2 symmetry for n_f = 4, explicit Boltzmann equations in Appendix C, detailed recast criteria in Appendix D, and a public UFO model file. However, the printed formula for the central cross section contains a sign/typographical error in the velocity scaling, the radiative stability of the mass degeneracy on which the mechanism rests is asserted but not demonstrated, and the claim that 3-to-2 processes are subdominant is not quantitatively supported. These issues are load-bearing and need to be addressed before the main conclusions can be accepted.","major_comments":[{"comment":"The printed Eqs. (3.1) and (C.7) place sqrt(x) in the numerator, while Eq. (C.6) and the surrounding text require 1/sqrt(x). With the printed factor the cross section increases at late times, reversing the claimed velocity suppression v proportional to 1/sqrt(x) that is the basis for evading indirect-detection and CMB bounds in the 1-10 GeV window. Please correct the factor in both equations and verify that the relic-density curve in Fig. 1 was computed with the corrected formula.","section":"Sec. 3.1, Eq. (3.1); Sec. C, Eq. (C.7)"},{"comment":"The statement that '2-to-2 processes are more efficient than the 3-to-2 ones, for all values of m_piD, fD and x under consideration' is not supported by any quantitative comparison in the text or appendices. Since Eqs. (C.16)-(C.17) include 3-to-2 terms and the relic abundance is then computed from the simplified Eq. (3.3), please provide the comparison (for example, a plot of the ratio of the relevant rates over the parameter space, or an analytic bound) that justifies dropping 3-to-2 processes.","section":"Sec. 3.1, bullet on 3-to-2 processes; Sec. C"},{"comment":"The exact degeneracy m_Q = m delta_alpha beta, called crucial in Sec. 2, is not radiatively stable: the portal (2.2) breaks SU(4)_V to SU(3) times U(1), so loop corrections involving kappa, X, and SM quarks generate different self-energies for Q4 and Q1-Q3. The paper acknowledges in Sec. 3.1 that 'a small mass splitting will be generated' and asserts Delta/m_piD much less than 1 'is expected', but no estimate or bound is given. Because the p-wave form of Eq. (3.1) and the indirect/CMB constraints in Fig. 1 rely on near-degenerate pi_DM and pi_tran, please provide a one-loop estimate of Delta/m_piD for the benchmarks used in the collider analysis (for example, kappa = 1 and m_X = 2 TeV) and identify the parameter region where Delta/m_piD much less than 1 holds.","section":"Sec. 2, Eq. (2.2); Sec. 3.1, after Eq. (3.1)"},{"comment":"The paper states that the emerging-jet recast follows the procedure of Ref. [56] with off-diagonal lifetimes fixed by Eq. (2.17), and explicitly calls this 'a less conservative choice than used in [18]', so the emerging-jet limits in Figs. 5-7 may be over-estimated. Since these limits are part of the central collider-reach claims, please either repeat the analysis with the conservative procedure of Ref. [18] or show quantitatively how much the limits change.","section":"Sec. 4.3, emerging jet search; Appendix D"}],"minor_comments":[{"comment":"The text in Sec. 3.4 states the direct detection lines correspond to c_tau = 1 cm and c_tau = 0.1 mm, while the Fig. 1 caption and Sec. 3.2 state c_tau = 10 cm and c_tau = 1 mm. These values should be made consistent.","section":"Sec. 3.4, caption of Fig. 1"},{"comment":"For f_D = 15 m_piD, the stated relation f_D approx Lambda_D/(4 pi) with Lambda_D = 40 m_piD gives f_D approx 3.2 m_piD, not 15 m_piD. Please clarify how Lambda_D is chosen for the f_D = 15 m_piD benchmarks.","section":"Sec. 4.2"},{"comment":"Reference [55], the ATLAS semi-visible jets search, is missing publication details; it should be completed before publication.","section":"Reference [55]"}],"recommendation":"major_revision","confidential_remarks":"The main risk to the paper's central claim is the radiatively induced mass splitting between stable and transient dark pions. If Delta/m_piD is not very small, the p-wave suppression and the resulting relic/indirect constraints change qualitatively. The authors should either compute this splitting or state clearly why it is protected. The typo in Eq. (3.1) is correctable, and the 3-to-2 dominance can presumably be shown with a short appendix plot. I do not see circularity concerns; the self-citations are appropriate and the core derivation is a first-principles effective-field-theory calculation."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This paper is worth a serious referee. It builds a genuinely new composite DM model: for n_f>=4, an unbroken dark flavor symmetry automatically stabilizes some pions, and the n_f=4 realization has an accidental Z2 that forces the stable pions to appear in pairs in the ChPT interactions. That combination is not in the cited literature, and it gives a minimal GeV-mass thermal DM candidate with clear LHC dark-shower signatures.\n\nThe group theory and the derivation of the velocity-suppressed co-annihilation cross section are careful. The relic abundance calculation is standard, and the collider study is thorough: four searches recast, honest discussion of efficiencies, and the code is public. The authors also flag their own softer choices: the emerging-jets recast is less conservative than Ref. [18], and hadronization parameters are unconstrained.\n\nThe soft spot is load-bearing. The entire low-mass window relies on degeneracy between stable and transient pions, which follows from the tree-level assumption m_Q = m delta. The portal breaks SU(4)_V radiatively, and the authors kick the issue to a footnote claiming Delta/m_pi << 1 'is expected'. They never compute or bound it. This matters: with kappa=1 and a coupling to top quarks, the one-loop correction to the dark quark mass is O(kappa^2/(16 pi^2) m_t log(m_X/m_t)) ~ 1-3 GeV, not small compared to m_pi in the 1-10 GeV window. If the splitting is that large, the p-wave suppression in Eq. (3.1) is lost and the relic, indirect-detection, and CMB constraints in Fig. 1 shift. The authors need to calculate the splitting, show a parameter region where it is small (e.g., no top coupling), or explain why this estimate is wrong.\n\nMinor issues: Eq. (3.1) and (C.7) have a sqrt(x) vs 1/sqrt(x) typo in intermediate expressions, contradicting the stated velocity scaling. The 3-to-2 subdominance is asserted without showing the promised check. Both are easily fixed.\n\nThe paper is for people working on composite DM, dark showers, and long-lived particle searches. It will be a useful benchmark regardless of the radiative-splitting question, but the central claim about the open GeV window is not yet proven. A good referee can give the authors a clear action list: compute or suppress the mass splitting, fix the typo, and show the 3-to-2 check. I would send it to review.","headline":"A creative composite-DM benchmark with a real radiative-stability hole in its central mechanism; worth refereeing seriously.","tokens_in":28446,"tokens_out":7600,"would_cite":true,"duration_ms":75483,"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":"An unbroken dark flavor symmetry makes a subset of dark pions stable dark matter, opening the GeV mass window and predicting mixed emerging and semi-visible jets at the LHC.","keywords":["composite dark matter","dark pions","dark flavor symmetry","dark QCD","impeded dark matter","emerging jets","semi-visible jets","long-lived particles"],"falsifier":"Measure the dark matter annihilation rate in dwarf galaxies with different velocity dispersions: the model predicts the rate scales linearly with relative velocity, so a velocity-independent rate would falsify the mechanism.","tokens_in":27342,"feed_emoji":"⚛️","tokens_out":11184,"duration_ms":88839,"temperature":0.7,"pith_summary":"This paper constructs a minimal composite dark matter model in which dark matter is a subset of the pions of a new confining gauge force, stable because of an unbroken flavor symmetry in the dark sector. The central claim is that with four or more dark quark flavors the stability of some dark pions is automatic, and the remaining unstable 'transient' pions control the relic abundance through co-annihilation that is velocity-suppressed by the degenerate pion masses. That suppression weakens indirect-detection and CMB constraints, making dark matter in the 1 to 10 GeV mass range viable where ordinary thermal WIMPs are excluded. Because dark matter is among the lightest dark-sector states, it is copiously produced in dark showers at colliders, predicting a mixture of emerging jets and semi-visible jets that the paper recasts existing LHC searches to constrain.","feed_headline":"Dark flavor symmetry stabilizes GeV-scale dark matter","feed_subtitle":"Dark QCD pions as DM evade indirect limits, yielding mixed emerging and semi-visible jets.","key_machinery":"The central object is the dark chiral Lagrangian for $SU(N_d)$ with $n_f$ dark quark flavors, with the flavor symmetry reduced by the portal coupling to a residual $G = SU(n_f-3) \\times U(1)$. The dark pion multiplet transforms under this residual symmetry, and the stable pions are those with nontrivial $G$ charges (for $n_f = 4$, a complex triplet under $SU(3)$). The mechanism that carries the argument is the velocity suppression of co-annihilation: a degenerate mass spectrum causes the thermally averaged cross section to scale as $1/\\sqrt{x}$, i.e. linearly with the velocity, which suppresses indirect detection and CMB signatures while still giving the correct relic density. The paper also uses an accidental $Z_2$ symmetry of the $n_f = 4$ chiral Lagrangian to show that stable pions appear only in pairs, ruling out odd-number 3-to-2 processes.","core_discovery":"The paper demonstrates that for $n_f \\geq 4$ dark quark flavors, the unbroken dark flavor symmetry $G = SU(n_f-3) \\times U(1)$ guarantees the stability of a subset of the dark pions, giving a dark matter candidate without imposing any additional discrete symmetry by hand. For $n_f = 4$ this yields six stable dark pions and nine transient ones. The relic abundance is set by co-annihilation of stable dark pions into transient dark pions, whose decay to Standard Model quarks is mediated by a heavy t-channel scalar portal. Because all dark pion masses are degenerate at tree level, the annihilation cross section carries a factor of the relative velocity, suppressing late-time annihilation and thereby evading gamma-ray and CMB bounds. The paper further shows that the stable pions appear only in even numbers in the chiral Lagrangian, so 3-to-2 processes do not disrupt the simple 2-to-2 freeze-out picture. The transient pions are naturally long-lived, and in the GeV dark matter mass window their decays produce a combination of semi-visible and emerging jets, which the paper uses to set limits on the mediator mass and dark pion mass.","pith_inferences":["The mechanism may generalise to other confining dark sectors: any dark sector where a flavoured pion multiplet is split into stable and unstable parts by a residual global symmetry could exhibit the same velocity-suppressed co-annihilation and long-lived decay signatures.","The accidental $Z_2$ protecting the stable pions for $n_f = 4$ is absent for larger $n_f$, so exploring $n_f = 5$ or 6 would test whether the dark matter stability and the simple 2-to-2 freeze-out picture persist beyond the specific case studied.","If the model accounts for the full relic abundance, the same parameters predict a specific flux of gamma rays from dark matter annihilation in dwarf galaxies at low velocities; a future measurement of that flux would either confirm or falsify the velocity-suppression prediction.","The paper's assumption of a single common decay width for transient pions could be relaxed in a detailed study, since off-diagonal pions have a slightly different lifetime, which would affect the emerging-jet sensitivity at intermediate lifetimes."],"forward_implications":["If the model is correct, dark matter in the few-GeV mass range is a natural thermal candidate, and the standard indirect-detection bounds do not apply because the annihilation rate today is velocity-suppressed.","Dark showers produced at hadron colliders would contain a significant fraction of missing energy even when the transient pions decay promptly, since stable dark matter pions are produced in the shower.","The collider signatures are a combination of semi-visible jets and emerging jets, so searches that merge these strategies are more sensitive than either search alone.","The relic abundance and the main detection cross sections depend only on the dark-sector parameters $m_{\\pi_D}$, $f_D$, $N_d$ and $n_f$, and not on the mediator mass or coupling, making the model predictive for a given dark pion mass.","For $n_f = 4$, the model selects a particular benchmark for future collider studies: mediator masses up to roughly 2.5 TeV are excluded by the combined searches for order-one portal couplings, and the remaining parameter space is testable at future colliders."],"supporting_citations":[{"why":"Defines the impeded dark matter paradigm that the model realizes, where velocity-suppressed annihilation avoids indirect detection bounds.","marker":"[1]"},{"why":"Prior composite dark matter study with $n_f=3$ that the paper contrasts; the model's key improvement is $n_f\\ge 4$ stability.","marker":"[8]"},{"why":"Original emerging jets proposal whose signatures the model predicts.","marker":"[17]"},{"why":"Original semi-visible jets proposal, the other key collider signature.","marker":"[20]"},{"why":"Semi-visible jets search recast to constrain the model's prompt-decay regime.","marker":"[55]"},{"why":"Emerging jets search with full Run 2 data, recast for long-lived dark pion lifetimes.","marker":"[56]"},{"why":"Jets plus missing energy search used for the long-lived-lifetime regime.","marker":"[59]"},{"why":"Four-jet search used for the prompt-decay regime.","marker":"[60]"},{"why":"Flavoured dark sector framework providing the portal structure and flavour constraints.","marker":"[22]"}],"fun_headline_variants":["Dark pions evade gamma-ray and CMB bounds","Stable dark pions from flavor symmetry","Co-annihilation sets dark matter abundance","Dark showers yield semi-visible and emerging jets"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The entire mechanism rests on the assumption that the dark quark mass matrix is exactly proportional to the identity, so that the diagonal dark flavor symmetry is preserved and all dark pion masses remain degenerate.","fun_headline_variants_meta":{"raw":{"variants":["Dark pions evade gamma-ray and CMB bounds","Stable dark pions from flavor symmetry","Co-annihilation sets dark matter abundance","Dark showers yield semi-visible and emerging jets"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000359,"raw_usage":{"total_tokens":1923,"prompt_tokens":902,"completion_tokens":1021,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":518,"completion_tokens_details":{"reasoning_tokens":963}},"tokens_in":518,"tokens_out":1021,"duration_ms":8417,"temperature":1.0,"reasoning_tokens":963,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T14:33:34.777423+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the dark matter annihilation rate in dwarf galaxies with different velocity dispersions: the model predicts the rate scales linearly with relative velocity, so a velocity-independent rate would falsify the mechanism.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Semi-visible jets search recast to constrain the model's prompt-decay regime."}],"review_version":1}