{"id":"ac29a9e5-d0b7-462e-af91-5d6b2088608a","arxiv_id":"2507.16555","paper_version":4,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"In coherent multi-ALP models, axion echo power scales linearly with the number of ALPs, while incoherent (random-phase) models yield signals no stronger than the single-ALP case.","lead":"This paper derives how axion 'echo' signals from dark matter change when there are many axion-like particles (ALPs) instead of one. It finds that coherently oscillating ALPs amplify the echo power linearly with their number, while random phases suppress it back to or below the single-ALP level.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The coherent equal-mass N-scaling is sound, but the mass-splitting amplification is not established: replacing the discrete ALP sum by a continuum before solving erases the discrete resonance condition m=2p, and the N=2 enhancement in Fig. 5 is unsupported.","rationale":"The paper's central coherent N-scaling is a clean, parameter-free consequence of adding N identical drives and is correctly normalized with the total DM density; recomputing the amplitude from Eqs. (3.20)-(3.22) confirms Eq. (3.23). The reader's identification of the LLN approximation as the weak point is correct, but I would strengthen it: the issue is not only finite-N fluctuations but the replacement of a discrete spectrum by a continuum before solving, which changes the resonance condition itself. The exact solution of the linear ODE is a sum over residues at mn=2p, and a continuum integral has a different asymptotic behavior, so the mass-splitting formulas (3.15), (3.18), (3.34)-(3.35) are not valid for finite N, including the N=2 claim in the abstract and Fig. 5. The incoherent scenario is less central and is consistent with the expectation of no N-enhancement. I recommend keeping the CONDITIONAL verdict because Eq. (3.23) stands while the derived mass-splitting amplification should be removed or replaced by an exact finite-N treatment; hence my read does not change the reader's verdict.","tokens_in":26549,"tokens_out":20812,"duration_ms":248293,"concrete_test":"Solve the exact N=2 version of Eq. (3.3) without the LLN replacement: take m1=mL, m2=3mL, equal couplings, amplitudes fixed by the total DM density (Eq. 3.22), set p=mL/2, and evaluate the time-averaged power from the forced-oscillator solution (A.12) at the experimental time t=R/v_perp. Compare with Eq. (3.35) and Fig. 5. If the ratio to the single-ALP power is not ~2.2 but at most the single-resonance value (approximately 0.1 for this density sharing), the mass-splitting amplification claim is refuted. More generally, repeat for uniformly spaced masses with N=10 and 30; if the power at the mL/2 line does not track the N-scaled formula, the continuum derivation in Sec. 3.1 and Appendix C is the cause.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The coherent equal-mass result PN=N P1 (Eq. 3.23) is mathematically sound. The problem is the additional mass-splitting amplification claimed in the abstract and Fig. 5. It is derived by replacing the discrete sum in Eq. (3.3) with the LLN continuum expression (3.11) and then solving the averaged equation. That interchange is not valid for resonant response. For a fixed incident frequency p, the source term for ALP n has a resonant Fourier component only when mn=2p; for N discrete masses this is a single spectral line, not a smooth density. The continuum replacement changes the pole structure of the solution: the exact finite-N sum has one linearly growing term from the matched ALP plus bounded off-resonance terms, whereas Appendix C's Taylor expansion produces secular t^2 and t^3 terms (Eq. C.31) from the averaged drive. Physically, for times larger than the inverse mass splitting the unmatched ALPs dephase and stop contributing, so the claimed ~1.1N enhancement at epsilon=1 (Eq. 3.37) is not credible for a discrete spectrum. The N=2 case is the clearest test: with masses mL and 3mL, at p=mL/2 the second ALP is off resonance by 2mL and cannot contribute to the same echo line, so the power cannot be 2.2 P1. This tension is visible in the paper itself, which notes around Eq. (3.6) that for distinct masses only one resonance dominates. Appendix B's N>4 criterion concerns fluctuations of the source sum at fixed t, not this resonance-pole failure.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper studies the back-scattered 'echo' radiation produced when a photon beam passes through a background of N axion-like particles (ALPs) coupled to the photon. After reviewing the single-ALP derivation, the authors generalize to multiple ALPs in two phase configurations: coherent (all ALP fields in phase) and incoherent (random phases). For the coherent equal-mass case they derive P_N = N P_{N=1} (Eq. 3.23), which improves coupling reach by sqrt(N). They further claim that small mass splittings produce an additional amplification, up to about 1.1N for epsilon=1 (Eq. 3.37), even for N=2, and present projected bounds in Fig. 5. For the incoherent case they find the N-dependence disappears and the signal is at most equal to the single-ALP one. The paper includes detailed appendices for the forced-oscillator solution, the large-N approximation, and the power calculation.","tokens_in":26962,"tokens_out":5552,"duration_ms":54435,"significance":"If the coherent N-scaling result holds, the paper provides a clean and potentially important effect: echo searches in multi-ALP models gain a factor sqrt(N) in coupling reach without any fitting to data. The derivation of P_N = N P_{N=1} is transparent and correct. The treatment of the incoherent case as a distinct scenario is also a useful warning that the N-enhancement is not generic. However, the additional mass-splitting amplification, which appears in the abstract and in Fig. 5, rests on a large-N continuum replacement applied to finite N. This is the paper's main advertised new quantitative claim, so its status determines the paper's overall impact. The appendices include a careful statement of the LLN validity condition (N>4 for uniform distributions at 10% accuracy), which is in tension with the N=2 application.","major_comments":[{"comment":"The 'additional amplification' claim for N=2 is derived from the continuum LLN replacement rather than from the finite-N discrete equations. The paper's own Eq. (3.6) states that for distinct masses only one resonance dominates; the exact finite-N solution for p=m_L/2 has only the m_L ALP growing linearly, while the off-resonant ALPs give bounded oscillatory terms. The continuum approximation replaces the discrete sum by an integral (Eq. 3.11) and produces secular t^2 and t^3 terms (Eq. C.31) that are artifacts of the averaging. Therefore the N=2 enhancement in Fig. 5 is unsupported, and the abstract's 'even for a N=2 case' claim should be withdrawn or supported by a direct N=2 calculation.","section":"Sec. 3.4, Eq. (3.37), Fig. 5"},{"comment":"The LLN condition (B.15) requires N > 4 for uniform distributions at the 10% precision level, and this condition concerns the fluctuation of the source sum, not the resonance-pole structure. Since the mass-splitting formula (3.18) is used for N=2 and N=10 in Fig. 5, the finite-N validity is not established. The paper needs to either restrict the mass-splitting amplification to N satisfying the LLN bound and to cases where the continuum resonance condition is physically justified, or compute the finite-N sum directly.","section":"Sec. 3.1 and Appendix B"},{"comment":"The limiting procedure used to recover the equal-mass result from the distribution formalism, lim_{epsilon->0} epsilon p(ma) = 1/2, is an ad hoc normalization condition rather than a well-defined distributional limit. While this does not affect the direct derivation of Eq. (3.23), it obscures the status of the general-distribution formula (3.18) and should be clarified or replaced by a direct delta-function treatment.","section":"Sec. 3.2, Eq. (3.25)"},{"comment":"The paper's own perturbative validity condition is epsilon m_L t = 1, giving t ~ 10^-11 sec for m_L ~ 10^-4 eV. The authors then replace t by R/v_perp, which is orders of magnitude larger, and conclude the echo should be detectable. However, the secular t^2 and t^3 terms used in Eq. (3.34) are derived under the assumption epsilon t << 1 (or at most epsilon m_L t ~ 1), so evaluating the power at R/v_perp is outside the domain of validity of the perturbative expansion. The projected bounds in Fig. 5 therefore require a non-perturbative treatment or a finite-N calculation for the relevant timescale.","section":"Sec. 3.4, Eq. (3.36)"}],"minor_comments":[{"comment":"There is a typo: 'strenghtening' should be 'strengthening'.","section":"Sec. 5"},{"comment":"The caption says 'mass ratio 3 corresponding to epsilon = 1'; since epsilon is defined as (m_M - m_L)/2, it would be clearer to state that epsilon/m_L = 1 for masses m_L and 3m_L.","section":"Fig. 5 caption"},{"comment":"Several references, e.g., [1], [5], and [13], lack complete page or article-number information; the bibliography should be standardized to the journal's style.","section":"References"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: the coherent N-scaling result (PN = N P1) is correct and cleanly derived. The secondary claim that small mass splittings add an extra amplification, even for N=2, does not survive the paper's own approximations. The large-N continuum replacement used to derive it is not valid for a discrete, resonant system.\n\nWhat the paper does well: the single-ALP review is compact, the forced-oscillator formulation makes the resonance condition p = ma/2 transparent, and the equal-mass coherent extension is genuinely new—the source term picks up a factor N and the power scales as N. The incoherent scenario, where random phases suppress the signal back to single-ALP levels, is a useful contrast and likely qualitatively correct even if the effective-drive treatment is heuristic. Appendix B's discussion of the Law of Large Numbers validity is honest and helpful.\n\nThe soft spot is Section 3.4. The mass-splitting enhancement is derived by replacing the discrete sum over N ALPs with a smooth integral over the mass distribution, then solving the averaged equation. For a driven oscillator at resonance, that interchange is not safe. At fixed photon momentum p, only the ALP with mn = 2p is on resonance; the off-resonant ALPs contribute bounded, oscillating terms. The continuum replacement smears the resonance and produces secular t^2 and t^3 terms that do not correspond to any physical drive. The stress-test note makes this concrete: for N=2 with masses mL and 3mL, at p = mL/2 the second ALP is off resonance by 2mL and cannot contribute to the same echo line, so the power cannot be 2.2 P1. The paper almost acknowledges this at Eq. (3.6), where it says that for distinct masses only one resonance dominates. Appendix B's N>4 criterion concerns fluctuations of the source sum, not this resonance-pole failure, so it does not rescue the N=2 claim.\n\nThe projected bounds in Fig. 5 also use a t-expansion beyond its perturbative validity: the paper gives the condition ϵ mL t = 1, but then replaces t with R/v⊥, which is orders of magnitude larger. The mass-splitting amplification and the associated sensitivity projections are therefore not reliable; they should be rederived treating the discrete spectrum exactly, or dropped.\n\nOverall: the central N-scaling result is solid, the incoherent suppression is a plausible qualitative contrast, and the mass-splitting claims are the weak part—unfortunately the part the abstract leads with. A referee would catch this. The paper deserves peer review because the core result is useful and the fix is local. For the reader: the axion-echo community gets a useful N-scaling result; the multi-ALP modeling community gets a caution about continuum approximations. I'd bring it to a reading group as an instructive case of a large-N replacement failing in a resonant system.","headline":"The coherent N-scaling result is solid and worth knowing; the mass-splitting boost for N=2 does not survive the paper's own approximations.","tokens_in":27440,"tokens_out":4574,"would_cite":true,"duration_ms":44164,"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":"The paper claims that N coherent axion-like particles amplify the axion echo power by a factor of N, with small mass splittings adding extra gain, while random phases suppress the signal below the single-ALP level.","keywords":["axion echo","axion-like particles","multi-ALP dark matter","photon-ALP coupling","stimulated axion decay","coherent and incoherent phases","large-N approximation","axion dark matter"],"falsifier":"Numerically integrate the exact $N=2$ forced-oscillator equation (Eq. 3.3) with masses $m_L^a$ and $m_L^a + 2\\epsilon$, equal couplings, $\\epsilon = m_L^a$, and sample at $m_L^a t \\sim 1$; if the echo power does not match $P_N^\\epsilon$ from Eq. (3.35) (about $1.1\\times 2P_{N=1}$), the continuum approximation fails at $N=2$ and the claimed enhancement is not established. A null echo search at the predicted boosted power would test the coherent multi-ALP claim directly.","tokens_in":26335,"feed_emoji":"📡","tokens_out":12360,"duration_ms":122479,"temperature":0.7,"pith_summary":"The paper develops the theory of axion echoes — returning photons produced when an outgoing beam stimulates the decay of axion-like-particle (ALP) dark matter — for the case where the dark matter consists of $N$ distinct ALPs rather than one. Its central claim is that if the $N$ fields oscillate coherently (same phase), the echo power is multiplied by $N$, as if there were a single ALP with a coupling $\\sqrt{N}$ times larger, and that a narrow spread of ALP masses adds a further amplification factor, noticeable even at $N=2$. In the opposite, incoherent case of random phases, the paper claims the echo power is suppressed to at most the single-ALP value and typically below it. This matters because multiple ALPs arise naturally in many beyond-standard-model constructions, so echo-search projections and the interpretation of null results depend on which scenario is realized.","feed_headline":"Coherent multi-ALP dark matter boosts echo power N-fold","feed_subtitle":"If dark-matter ALPs oscillate in phase, echo searches get N times more signal; random phases erase the gain.","key_machinery":"The load-bearing object is the forced-oscillator equation for the first-order photon perturbation, $(\\partial_t^2 + p^2)\\mathbf{A}_p^1 = -i\\sum_n \\mathbf{B}_{kp}^n [e^{i(m_n^a-p)t} + e^{-i(m_n^a+p)t}]$, together with the resonance condition $p = m_n^a/2$, at which the amplitude grows linearly in time and the echo wave propagates backward. The multi-ALP argument converts the sum over $N$ fields into an integral over smooth mass and coupling distributions (Eq. 3.11), justified in the large-$N$ limit by the Law of Large Numbers, and then expands the mass integral in the small splitting $\\epsilon = (m_M^a - m_L^a)/2$, producing the dimensionless coefficients $a_1$, $a_2$ and the amplification factor $Z(\\epsilon,t)$. In the incoherent treatment the same sum is replaced by its root-mean-square amplitude, which removes the $N$ enhancement. These replacements — discrete sum to continuum integral, coherent summation versus RMS — are what carry the paper's results.","core_discovery":"On the paper's own terms, the discovery is a scaling law for echo power: for $N$ coherent ALPs with equal masses and couplings, the first-order echo field amplitude is $N$ times the single-ALP amplitude, giving $P_N = N P_{N=1} = N g_{a\\gamma\\gamma}^2 (t/16) \\rho \\, dP_0/d\\nu\\big|_{k=m_a/2}$ (Eq. 3.23), equivalent to a single ALP with coupling $\\sqrt{N}\\,g_{a\\gamma\\gamma}$. When the masses are drawn from a narrow distribution with splitting $\\epsilon$, a large-$N$ continuum replacement plus Taylor expansion in $\\epsilon$ yields an additional factor $Z(\\epsilon,t)>1$ in the power, bounded near $1.1N P_{N=1}$ at maximal $\\epsilon=1$ with $m_a^L t\\sim 1$; the paper therefore claims stronger projected bounds even for $N=2$ (Fig. 5). For incoherent phases, replacing the random drive by its root-mean-square value removes the $N$ enhancement entirely: $P_N=f_g P_{N=1}$ with $f_g\\le 1$, and $f_g=1/3$ for a uniform coupling distribution, so the observable signal is at best that of a single ALP and often weaker.","pith_inferences":["Beyond the paper: measuring echo power at two closely spaced outgoing frequencies should reveal the mass-distribution-dependent factor $Z(\\epsilon,t)$ in the coherent case but no such structure in the incoherent case, giving an observational way to distinguish the two scenarios.","Beyond the paper: a direct numerical solution of the exact $N=2$ and $N=3$ forced-oscillator systems (without the continuum replacement) at $\\epsilon=1$ would test whether the 'even $N=2$' amplification survives finite-$N$ fluctuations, which Appendix B's accuracy bound leaves open.","Beyond the paper: because the mass-splitting amplification depends on $p(m_L^a)$, $p'(m_L^a)$ and $p''(m_L^a)$, a sufficiently precise measurement of echo power versus frequency could in principle reconstruct moments of the ALP mass distribution, turning echo searches into a spectroscopy tool.","Beyond the paper: the incoherent suppression implies that ALP models with random phases, such as fields produced at different epochs, would be systematically harder to detect by the echo method than single-ALP dark matter of the same total density; null results should therefore carry a model-dependent suppression factor."],"forward_implications":["If coherent multi-ALP dark matter is realized, echo experiments effectively search for a coupling $\\sqrt{N}$ larger than in the single-ALP case, so projected bounds on $g_{a\\gamma\\gamma}$ tighten by a factor $\\sim\\sqrt{N}$ for fixed signal power.","If the ALP masses are spread over a narrow band, the extra factor $Z(\\epsilon,t)$ adds up to roughly 10% amplification at maximal $\\epsilon=1$, enough for even an $N=2$ model to give stronger projected constraints than the single-ALP baseline (Fig. 5).","For random ALP phases, the echo power is at most equal to the single-ALP power (and one-third of it for uniform couplings), so a null search interpreted under the coherent assumption would claim sensitivity to multi-ALP dark matter that the incoherent scenario does not actually provide.","For ALPs with well-separated masses, such as Kaluza-Klein towers, only one resonance is active at a time and the signal matches the single-ALP case, so the $N$ amplification requires near-degenerate masses.","The mass range $m_L^a \\in [2.5\\times10^{-7}, 2.5\\times10^{-3}]$ eV lies in the atmospheric transparency window, so multi-ALP echo searches would yield bounds complementary to haloscope-type dark-matter axion searches."],"supporting_citations":[{"why":"Establishes the stimulated photon enhancement at photon momentum equal to half the axion mass, the physical basis of the echo.","marker":"[25]"},{"why":"Supplies the single-ALP echo derivation and the detector/projection setup that every multi-ALP result extends and compares against.","marker":"[26]"},{"why":"Provides the detailed single-ALP echo analysis whose power and timescale estimates anchor the multi-ALP treatment.","marker":"[27]"},{"why":"Supplies the string-landscape logarithmic mass distribution used to argue that the uniform small-splitting approximation holds.","marker":"[40]"},{"why":"Gives the clockwork mass spectrum whose near-uniform small splittings the coherent calculation is designed to model.","marker":"[43]"},{"why":"Motivates multi-ALP dark matter and supplies the $N=30$ example used in the projected bounds.","marker":"[45]"},{"why":"Fixes the isothermal dark-matter density profile used in all projected sensitivity curves.","marker":"[49]"}],"fun_headline_variants":["Coherent ALPs amplify echo power N-fold","Incoherent ALPs erase N-fold echo gain","Echo power scales N times with coherent ALP stacks","Coherent multi-ALP ensembles N-fold echo signal","ALP echo power: coherent phases multiply, random phases divide"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The paper's extra mass-splitting amplification rests on treating the $N$ ALPs as a smooth continuum distribution, an approximation its own appendix shows needs more than four ALPs for ten-percent accuracy, even though the $N=2$ amplification is computed from that large-$N$ formula.","fun_headline_variants_meta":{"raw":{"variants":["Coherent ALPs amplify echo power N-fold","Incoherent ALPs erase N-fold echo gain","Echo power scales N times with coherent ALP stacks","Coherent multi-ALP ensembles N-fold echo signal","ALP echo power: coherent phases multiply, random phases divide"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001252,"raw_usage":{"total_tokens":5152,"prompt_tokens":984,"completion_tokens":4168,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":600,"completion_tokens_details":{"reasoning_tokens":4089}},"tokens_in":600,"tokens_out":4168,"duration_ms":30185,"temperature":1.0,"reasoning_tokens":4089,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T15:08:43.320708+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Numerically integrate the exact $N=2$ forced-oscillator equation (Eq. 3.3) with masses $m_L^a$ and $m_L^a + 2\\epsilon$, equal couplings, $\\epsilon = m_L^a$, and sample at $m_L^a t \\sim 1$; if the echo power does not match $P_N^\\epsilon$ from Eq. (3.35) (about $1.1\\times 2P_{N=1}$), the continuum approximation fails at $N=2$ and the claimed enhancement is not established. A null echo search at the predicted boosted power would test the coherent multi-ALP claim directly.","supporting_citations":[{"cited_title":"Arza, Photon enhancement in a homogeneous axion dark matter background , Eur","cited_arxiv_id":null,"evidence_quote":"Establishes the stimulated photon enhancement at photon momentum equal to half the axion mass, the physical basis of the echo."},{"cited_title":"Arza and P","cited_arxiv_id":null,"evidence_quote":"Supplies the single-ALP echo derivation and the detector/projection setup that every multi-ALP result extends and compares against."},{"cited_title":"Arza and E","cited_arxiv_id":null,"evidence_quote":"Provides the detailed single-ALP echo analysis whose power and timescale estimates anchor the multi-ALP treatment."},{"cited_title":"Broeckel, M","cited_arxiv_id":null,"evidence_quote":"Supplies the string-landscape logarithmic mass distribution used to argue that the uniform small-splitting approximation holds."},{"cited_title":"Giudice and M","cited_arxiv_id":null,"evidence_quote":"Gives the clockwork mass spectrum whose near-uniform small splittings the coherent calculation is designed to model."},{"cited_title":"Chadha-Day, J","cited_arxiv_id":null,"evidence_quote":"Motivates multi-ALP dark matter and supplies the $N=30$ example used in the projected bounds."}],"review_version":1}