{"id":"c494b620-4763-4bb6-8291-8fe2423ce9b5","arxiv_id":"1202.0283","paper_version":2,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"Whether dark matter annihilation recouples after freeze-out depends on whether the interaction is even or odd under charge conjugation, not just on whether oscillations have begun.","lead":"The paper derives the density-matrix Boltzmann equations for dark matter particle-antiparticle oscillations from nonequilibrium field theory, and shows that interactions odd under charge conjugation block annihilation recoupling while even ones block decoherence. It corrects an earlier treatment and moves indirect-detection bounds on asymmetric dark matter by many orders of magnitude.","discovery_kind":"first_principles","skeptic_critique":{"model":"selfhosted/deepseek-v4.1-flash","headline":"The paper's own thermal-decoherence estimate implies flavor-sensitive annihilation should restart around x ~ 375 for the Fig. 2 benchmark, yet Fig. 2 bottom left shows a flat density out to x ~ 1000; the figure does not reflect the acknowledged limitation.","rationale":"The paper's central claim is sound and correctly scoped. The reader identified the factorization ansatz as the weakest assumption, which is correct. I sharpen it: the figure's flat curve is in tension with the paper's own decoherence estimate for the benchmark parameters, and the paper should annotate the figure accordingly. This does not change the verdict because the claim is explicitly limited to the onset of coherent oscillations, and the paper states that thermal decoherence is not captured by the integrated equation. But the figure could mislead a reader into thinking flavor-sensitive annihilation never recouples for these parameters, when the paper's own estimate says it should recouple around x ~ 375.","tokens_in":22385,"tokens_out":9905,"duration_ms":200588,"concrete_test":"Overlay tau_dec(x) = 1/(delta m v^2(x)) with v^2(x) ~ x^{-1} on Fig. 2 bottom left and identify the x where tau_dec = 1/H. For the benchmark parameters this is x ~ 375. If the figure's flat Sigma extends beyond this x, the curve is not physical there. Alternatively, solve Eq. (6) directly for F_k with omega_osc(k) = 2 delta m (1 - k^2/(2 m_X^2)) and check whether Sigma depletes around x ~ 375.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is correctly scoped (flavor-sensitive annihilation does not recouple when coherent oscillations begin), and the paper acknowledges that thermal decoherence from O(k^2/m_X^2) corrections will eventually restart annihilation (Sec. III A, Eqs. 24-25). But the numerical figure (Fig. 2 bottom left, kappa=0) shows Sigma flat out to x ~ 1000, which is inconsistent with the paper's own estimate tau_dec ~ 1/(delta m v^2). For m_X = 300 GeV, delta m = 10^-7 eV, the crossover tau_dec ~ 1/H occurs around x ~ 375, and the annihilation rate at that point is a few times H, so the density should visibly deplete. The figure therefore overstates the persistence of the frozen-out density. The paper should either overlay the decoherence estimate on the figure or state explicitly that the flat curve is not physical beyond x ~ 375. This is a presentation issue, not a flaw in the central claim, but it matters because the figure is the primary numerical evidence for the flavor-sensitive case.","agreement_with_reader":"partial"},"referee_report":{"model":"selfhosted/deepseek-v4.1-flash","summary":"The manuscript studies particle-antiparticle oscillations of asymmetric dark matter (ADM) generated by a small Majorana-type mass term, and derives density-matrix Boltzmann equations from nonequilibrium field theory. The central result is that the matrix structure of the collision term depends on the C-parity of the DM interaction: flavor-sensitive (C-odd) annihilation couples only to det(Y) and therefore does not recouple when coherent oscillations begin, while flavor-blind (C-even) scattering does not cause decoherence. The paper supports this with an analytic argument in Sec. III D and with numerical solutions in Figs. 2 and 3, and discusses implications for indirect detection.","tokens_in":22560,"tokens_out":17917,"duration_ms":348101,"significance":"The paper provides a careful field-theory derivation of the density-matrix equations for oscillating ADM and identifies a qualitatively important distinction between flavor-blind and flavor-sensitive interactions that was not captured in earlier treatments. The analytic proof in Sec. III D is clean and the results have clear implications for relic-density calculations and indirect-detection bounds. The work is a valuable contribution to the ADM literature and has been widely cited.","major_comments":[{"comment":"The flat density curve out to x ~ 1000 is a solution of Eq. (22), which by construction neglects the O(k^2/m_X^2) corrections to omega_osc that the paper itself identifies in Sec. III A (Eqs. 24-25) as the source of thermal decoherence. For the benchmark parameters m_X = 300 GeV and delta m = 10^-7 eV, the paper's own estimate tau_dec ~ 1/(delta m v^2) implies that decoherence and renewed annihilation should set in around x ~ 300-1000, so the figure overstates the persistence of the frozen-out density. The caption and the accompanying statement in Sec. III C that Omega_DM = m_X eta_DM s_0/rho_c is fixed by the initial asymmetry for the flavor-sensitive kappa = 0 case should be qualified, or the decoherence scale should be overlaid on the figure.","section":"Sec. III C, Fig. 2 bottom left (and Fig. 3 bottom left)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The ansatz F_k = exp(-omega_k/T) n_eq assumes that all density-matrix elements share a common Maxwell-Boltzmann momentum dependence. The paper's justification -- that omega_osc is approximately k-independent in the nonrelativistic limit -- is a plausibility argument rather than a derivation, and the neglected O(k^2/m_X^2) corrections are precisely the ones that drive thermal decoherence. It would be useful to state explicitly that Eq. (18) is an approximation and to estimate its accuracy, or to show that the collision term does not generate a different momentum dependence for the off-diagonal elements.","section":"Sec. II B, Eq. (18)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The paper states that for cases other than flavor-blind annihilation without scattering and flavor-sensitive annihilation with scattering, its results differ from Ref. [22] by an order of magnitude in the relic density. This is a strong quantitative claim about another work, but no direct comparison (figure or table) is shown. A side-by-side comparison of the relic densities obtained with the two collision-term structures would make the claim verifiable and would clarify exactly which matrix-structure difference is responsible.","section":"Sec. III C","authors_acknowledge":null,"authors_acknowledge_p":null}],"minor_comments":[{"comment":"The statement that flavor-sensitive annihilation 'does not recouple when coherent oscillations begin' should be qualified, since the body of the paper (Sec. III A) shows that thermal decoherence or flavor-sensitive scattering will eventually restart annihilation. A phrase such as 'does not recouple until coherence is broken' would be more precise.","section":"Abstract","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The CMB line appears to contain a typo. With T in TeV, the prefactor should be 10^8 rather than 10^-16; equivalently, the formula should use (T/eV)^2 with prefactor 10^-16. As written, the CMB velocity estimate is off by 24 orders of magnitude.","section":"Eq. (33)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The claim that flavor-blind scattering does not lead to decoherence follows from O_+ = identity in Eq. (10) and the assumption f_p = f_bar_p. It would be useful to state these conditions explicitly, since a more general flavor-blind interaction (e.g., with derivative couplings or a different momentum dependence) could in principle generate decoherence.","section":"Sec. III B","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The scattering term includes Xf <-> Xf and X_bar f <-> X_bar f but not oscillation-induced processes such as Xf <-> X_bar f. It would be useful to state whether these are negligible (e.g., suppressed by delta m/omega) or whether they should be included in a more complete treatment.","section":"Eq. (10)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The notation for O_pm and M_pm could be clarified. The subscript pm on O refers to the diagonal structure (1, pm 1), while the subscript pm on M refers to the C-parity of the amplitude; a sentence spelling this out would help the reader.","section":"Eq. (16)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The kappa = 0 curves should be labeled explicitly as solutions of Eq. (22) that neglect thermal decoherence, so that the reader does not interpret the flat curves as physical at arbitrarily late times.","section":"Fig. 2 and Fig. 3 captions","authors_acknowledge":null,"authors_acknowledge_p":null}],"recommendation":"minor_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"selfhosted/deepseek-v4.1-flash","letter":"Two things worth knowing. First, the central claim is right and it's a real correction to the literature. The collision term's matrix structure depends on the C-parity of the DM interaction: flavor-sensitive annihilation couples only to det(Y), and since det([H0, Y]) = 0, oscillations can't source it. That's a clean algebraic argument (Sec. III D), and it means flavor-sensitive annihilation doesn't recouple when coherent oscillations begin. The companion claim — flavor-blind scattering doesn't decohere because O_+ is the identity — is equally clean. The CTP derivation in the appendix is standard Kadanoff-Baym done carefully, and the paper is honest that it agrees with Ref. [22] in two of the four cases and differs qualitatively in the others. Second, the soft spot is the numerical figure for the flavor-sensitive, no-scattering case (Fig. 2 bottom left). The paper's own thermal-decoherence estimate (Eqs. 24-25) says the O(k^2/m_X^2) corrections restart annihilation on a timescale tau_dec ~ 1/(delta m v^2), which for the Fig. 2 benchmark is well inside the plotted range. The figure shows a flat density out to x~1000 because it solves Eq. (22), which the paper explicitly says neglects this effect. That's technically consistent, but the paper then states the relic density is 'fixed by the initial asymmetry' for this case without qualifying that the full physics would deplete it. A reader could take the flat curve as the physical answer. The fix is a sentence in the caption or text saying the curve is the solution of Eq. (22) only, and marking where the approximation breaks down. Other soft spots, in proportion: the flavor-blind scattering result relies on the contact-interaction form and on neglecting the medium-induced mass shift, which the paper flags in footnote 6 but doesn't quantify. The formalism citation to Ref. [27] is to a forthcoming paper, though the appendix is self-contained enough that this doesn't block reimplementation. The numerical benchmarks are illustrative, not fitted, so no circularity. Who should read it: anyone working on asymmetric DM, oscillating DM, or density-matrix Boltzmann equations in cosmology. The flavor-sensitive/flavor-blind distinction is a useful organizing principle that will show up in later work. Recommendation: send it out. The central claim is correct and the correction to Ref. [22] matters. The figure issue is a presentation fix, not a reason to hold it up.","headline":"The central claim is a clean, correct algebraic result that fixes a specific earlier treatment; the main soft spot is a numerical figure that doesn't reflect the paper's own decoherence estimate.","tokens_in":23235,"tokens_out":6535,"would_cite":true,"duration_ms":118964,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"pith_extraction":{"msc":[],"pacs":["95.35.+d","11.10.Wx"],"model":"selfhosted/deepseek-v4.1-flash","headline":"Whether dark matter annihilation restarts after freeze-out depends on how the dark matter interaction behaves under charge conjugation, and the paper derives the density-matrix equations that make this distinction unavoidable.","keywords":["asymmetric dark matter","dark matter oscillations","density matrix equations","nonequilibrium field theory","flavor oscillations","decoherence","freeze-out","charge conjugation parity"],"falsifier":"Solve Eq. (6) for the full momentum-dependent $F_k$ without the factorization ansatz, keeping the $k^2/m_X^2$ corrections to $omega_{osc}(k)$, and check whether flavor-sensitive annihilation stays suppressed or instead recouples on the thermal decoherence time $tau_{dec} sim 1/(delta m, v^2)$; a recoupling at that scale would undercut the claim that flavor-sensitive annihilation does not restart when oscillations begin.","tokens_in":23672,"feed_emoji":"🌑","tokens_out":2851,"duration_ms":60529,"temperature":0.7,"pith_summary":"Asymmetric dark matter assumes a stored particle-antiparticle imbalance set by a conserved charge, so that annihilation has already wiped out the symmetric part of the population long before today. A tiny charge-violating Majorana mass makes the dark particle oscillate into its own antiparticle, and the natural expectation is that this oscillation refills the antiparticle population and switches annihilation back on. This paper re-derives the governing equations from nonequilibrium field theory rather than borrowing the neutrino-oscillation template, and finds that the answer depends on a property previous treatments ignored: whether the dark matter bilinear coupling to lighter states is even or odd under charge conjugation. For the odd, or flavor-sensitive, case — annihilation through a new vector boson — the annihilation term couples only to $Y_{11}Y_{22}-Y_{12}Y_{21}$, and because the oscillation commutator generates no determinant, coherent oscillations do not restart annihilation at all. The paper also shows the mirror statement for scattering: flavor-blind scattering drops out of the collision term entirely and cannot destroy coherence, so only flavor-sensitive scattering can decohere the oscillating state.","feed_headline":"Dark matter's C-parity decides if annihilation restarts","feed_subtitle":"Oscillating into its own antiparticle does not revive annihilation when the interaction is flavor-sensitive.","key_machinery":"The load-bearing object is the integrated density-matrix Boltzmann equation, $Y'(x) = -frac{i}{Hx}[H_0,Y] - frac{Gamma_pm}{2Hx}[O_pm,[O_pm,Y]] - frac{s langle sigma v rangle_pm}{Hx}left(frac{1}{2}{Y, O_pm Ybar O_pm} - Y_{eq}^2right)$, obtained by projecting the Kadanoff-Baym equations of the closed-time-path formalism onto a number-density matrix and then integrating over momenta with a Maxwell-Boltzmann factorization ansatz. Its work is to separate three effects that ordinary Boltzmann equations fuse together: the coherent rotation generated by the Majorana splitting $delta m$, the decohering effect of elastic scattering, and the depletion from annihilation, each of which carries a different matrix structure in the two-flavor space spanned by particle and antiparticle. The charge-conjugation parity of the interaction, encoded in $O_pm$, is what decides whether each term acts on the diagonal populations, on the off-diagonal coherences, or on the determinant alone.","core_discovery":"The paper's central claim is that the matrix structure of the collision term in the dark matter density-matrix Boltzmann equation is not universal but is fixed by the transformation property of the interaction under $X$ to $X^C$. Writing the amplitude matrices as $C_a = M(X X^C to f f) O_pm$ with $O_pm = mathrm{diag}(1, pm 1)$, the flavor-blind ($+$) annihilation term is $frac{1}{2}{Y, O_+ Ybar O_+}$ while the flavor-sensitive ($-$) term reduces to $mathrm{diag}(Y_{11}Y_{22}-Y_{12}Y_{21},, Y_{11}Y_{22}-Y_{12}Y_{21})$, i.e. it depends only on $det Y$. Since $det([H_0,Y])=0$, the oscillation term cannot source flavor-sensitive annihilation, so a dark sector that froze out as a pure particle state and is now oscillating coherently stays frozen out. The authors back this with an analytic solution of the flavor-sensitive system, showing $Sigma = Y_{11}+Y_{22}$ remains pinned at the asymmetry $eta_{DM}$ while $Delta$ and $Xi$ oscillate as $cos(delta m/H)$ and $sin(delta m/H)$. The companion result is that the scattering double commutator $frac{Gamma_pm}{2}[O_pm,[O_pm,Y]]$ vanishes identically for $O_+$, so flavor-blind scattering causes no decoherence, whereas flavor-sensitive scattering damps $Y_{12}$ and $Y_{21}$ and, when $Gamma_- gg omega_{osc}$, suppresses oscillation through the quantum Zeno effect.","pith_inferences":["If flavor-sensitive annihilation truly stays off while coherence persists, the late-time annihilation rate in such models is controlled not by the oscillation frequency but by whatever breaks coherence, which makes the present-day annihilation signal a direct probe of the decoherence channel rather than of the Majorana mass.","The paper's own momentum-spread argument, $tau_{dec} sim 1/(delta m, v^2)$, suggests the flavor-sensitive suppression is temporary in any realistic thermal ensemble; a full momentum-resolved solution would then translate into a sharp prediction for the epoch at which the annihilation signal switches on, testable against CMB energy-injection limits.","The same even/odd bookkeeping should apply to any two-state system coupled to a bath through a bilinear whose parity under the state-swapping symmetry is definite, so the result plausibly carries over to oscillating scalar dark matter, to neutron-antineutron-like oscillations, and to any hidden-sector particle mixing with its own antiparticle."],"forward_implications":["For flavor-sensitive annihilation with negligible scattering, the relic abundance is fixed by the initial asymmetry alone, $Omega_{DM} = m_X eta_{DM} s_0/rho_c$, so weak-scale asymmetric dark matter cannot be washed down to the observed density by this mechanism.","Indirect-detection bounds derived by requiring the oscillation time $delta m^{-1}$ to exceed the BBN, CMB, and present-epoch annihilation constraints apply only to flavor-blind annihilation; for the flavor-sensitive case the authors estimate the bounds weaken by roughly six to sixteen orders of magnitude.","Flavor-sensitive scattering delays the onset of oscillation through the quantum Zeno effect until $omega_{osc} > Gamma_-$, and only after that does decoherence-driven annihilation deplete the density, which the numerical examples show can still give an order-of-magnitude reduction.","The framework widens asymmetric dark matter model building by making the relic density depend on interaction type rather than only on cross section and asymmetry, and it supplies a natural bridge between symmetric WIMP freeze-out and asymmetric freeze-out."],"supporting_citations":[{"why":"The prior density-matrix treatment of oscillating asymmetric dark matter that this paper re-derives and corrects, and the source of the numerical benchmark cases it reproduces.","marker":"[22]"},{"why":"The earlier heuristic treatment of dark matter oscillations without coherence, and the origin of the indirect-detection bounds the paper argues apply only to flavor-blind annihilation.","marker":"[21]"},{"why":"The neutrino-oscillation density-matrix equations in the early Universe that supply the template the authors replace with a first-principles derivation.","marker":"[23]"},{"why":"The closed-time-path derivation of Boltzmann equations that the appendix follows in obtaining the collision term.","marker":"[26]"},{"why":"The fermionic density-matrix formalism in the CTP approach on which the Wigner-transformed Green's function construction rests.","marker":"[27]"},{"why":"The asymmetric dark matter framework whose freeze-out treatment this paper generalizes to include particle-antiparticle coherence.","marker":"[5]"}],"fun_headline_variants":["Flavor-sensitive dark matter annihilation stays off during oscillation","Only the flavor-blind dark matter annihilation term can recouple","Det Y alone sets the flavor-sensitive dark matter annihilation rate","Flavor-blind dark matter scattering leaves oscillations coherent","Oscillating dark matter stays frozen out under flavor-sensitive annihilation"],"cache_read_input_tokens":0,"weakest_assumption_plain":"The derivation integrates the momentum-resolved equations using a factorization ansatz in which every element of the density matrix shares one Maxwell-Boltzmann momentum profile, which is precisely what discards the $O(k^2/m_X^2)$ spread in oscillation frequency that the paper elsewhere identifies as the mechanism by which flavor-sensitive annihilation eventually switches back on.","fun_headline_variants_meta":{"raw":{"variants":["Flavor-sensitive dark matter annihilation stays off during oscillation","Only the flavor-blind dark matter annihilation term can recouple","Det Y alone sets the flavor-sensitive dark matter annihilation rate","Flavor-blind dark matter scattering leaves oscillations coherent","Oscillating dark matter stays frozen out under flavor-sensitive annihilation"]},"model":"selfhosted/deepseek-v4.1-flash","effort":"low","attempts":1,"cost_usd":0.0,"raw_usage":{"total_tokens":1434,"prompt_tokens":1057,"completion_tokens":377,"prompt_tokens_details":null,"completion_tokens_details":{"reasoning_tokens":310}},"tokens_in":1057,"tokens_out":377,"duration_ms":7817,"image_pages":[],"temperature":1.0,"reasoning_budget":50,"reasoning_tokens":310,"transport_provider":"selfhosted","cache_read_input_tokens":0,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-11-10T15:27:53.368733+00:00","model_set":{"reader":"selfhosted/deepseek-v4.1-flash"},"falsifier":"Solve Eq. (6) for the full momentum-dependent $F_k$ without the factorization ansatz, keeping the $k^2/m_X^2$ corrections to $omega_{osc}(k)$, and check whether flavor-sensitive annihilation stays suppressed or instead recouples on the thermal decoherence time $tau_{dec} sim 1/(delta m, v^2)$; a recoupling at that scale would undercut the claim that flavor-sensitive annihilation does not restart when oscillations begin.","supporting_citations":[{"cited_title":"Consequences of DM/antiDM Oscillations for Asymmetric WIMP Dark Matter","cited_arxiv_id":"1110.3809","evidence_quote":"The prior density-matrix treatment of oscillating asymmetric dark matter that this paper re-derives and corrects, and the source of the numerical benchmark cases it reproduces."},{"cited_title":"Regenerating a Symmetry in Asymmetric Dark Matter","cited_arxiv_id":"1109.2164","evidence_quote":"The earlier heuristic treatment of dark matter oscillations without coherence, and the origin of the indirect-detection bounds the paper argues apply only to flavor-blind annihilation."},{"cited_title":"Sov. J. Nucl. Phys. 33, 700-706 (1981). G. Sigl, G. Raffelt, Nucl. Phys. B406, 423-451","cited_arxiv_id":null,"evidence_quote":"The neutrino-oscillation density-matrix equations in the early Universe that supply the template the authors replace with a first-principles derivation."},{"cited_title":"Flavored Quantum Boltzmann Equations","cited_arxiv_id":"0912.3523","evidence_quote":"The closed-time-path derivation of Boltzmann equations that the appendix follows in obtaining the collision term."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"The fermionic density-matrix formalism in the CTP approach on which the Wigner-transformed Green's function construction rests."},{"cited_title":"Phys. Rev. D 79, 115016","cited_arxiv_id":null,"evidence_quote":"The asymmetric dark matter framework whose freeze-out treatment this paper generalizes to include particle-antiparticle coherence."}],"review_version":1}