{"id":"9267d548-3fd0-4abb-b7bb-2f3af7a084ff","arxiv_id":"2412.06301","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Projected EicC searches could probe dark photons from about 1 MeV to 3 GeV and photon-coupled axion-like particles from 0.1 to 5 GeV, covering parameter space not yet excluded by other experiments.","lead":"This paper computes how well a proposed Chinese electron-ion collider could spot two types of hypothetical lightweight particles: dark photons and axion-like particles. It argues the machine would fill gaps in current experimental coverage, especially for particles that travel a short distance before decaying.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Displaced dark-photon reach assumes N_A'=10/100 signal events with no SM background model; the gap-filling claim is not yet demonstrated.","rationale":"The reader's weakest assumption emphasizes unpublished detector parameters and vertex-resolution geometry. Those are real limitations, but they shift the reach parametrically: varying L1 from 10 um to 100 um or L2 from 1 m to 0.14 m changes the displaced-decay probability smoothly and can be bracketed. The more load-bearing issue is the absence of any SM background model for the displaced dark-photon search. The central claim that the EicC fills a discovery gap depends on N_A' = 10 signal events being a meaningful sensitivity threshold; if even a handful of SM events populate the same vertex and invariant-mass windows, the projected boundary moves or disappears. The paper acknowledges this dependence but supplies no calculation, and the single sentence defending N_A' = 100 does not address the N_A' = 10 boundary. This is an addressable omission rather than a fatal flaw, so the reader's CONDITIONAL verdict remains appropriate; the condition should explicitly include a background estimate for the displaced search.","tokens_in":14964,"tokens_out":10538,"duration_ms":127609,"concrete_test":"Generate SM coherent e-Pb events for 10 fb^-1 and count gamma to e+e- conversions and pi0/eta Dalitz decays whose two tracks satisfy Eq. (2.6), have a reconstructed vertex between L1 = 10 um and L2 = 1 m (and also L2 = 0.14 m), and have a dilepton invariant mass within the detector mass resolution around a grid of mA' values such as 0.01, 0.1, and 1 GeV. If the surviving background exceeds roughly 1 event in any mass bin, the N_A' = 10 sensitivity boundary in Fig. 5 is not a valid projection and the gap-filling claim requires revision; if the background is below 0.1 events per bin, the background-free assumption survives.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 2.1 defines the displaced-search sensitivity by requiring N_A' = 10 or 100 signal events after Eq. (2.7), and the text explicitly says the result depends on assumptions about the background without estimating one. The abstract's claim that the EicC can comprehensively probe previously unexplored parameter space relies on these thresholds being effectively background-free. In an e-Pb collider with 10 fb^-1, coherent photoproduction of pi0/eta, Dalitz decays, and photon conversions in the 14 cm to 1 m tracking/decay volume can produce e+e- pairs with displaced vertices; no invariant-mass window, vertex-quality cut, or conversion veto is specified. The sentence 'Even if there exist some backgrounds, the bounds for signal event number N_A'=100 can still uniquely cover large areas of unexplored parameter space' is an assertion, not a calculation, and it does not justify the N_A'=10 curve that defines the lower edge of the claimed reach.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper presents projected sensitivities for the proposed Electron-Ion Collider in China (EicC) to two BSM particles: dark photons produced by electron bremsstrahlung in coherent e-Pb scattering, and photon-coupled ALPs produced by photon fusion. For dark photons, the displaced-vertex search uses Eq. (2.7) with vertex resolution L1 = 10 micrometres and decay lengths L2 = 0.14 m or 1 m, requiring 10 or 100 signal events at 10 fb^-1 (plus 3 events at 100 fb^-1) and considering only leptonic decays. For ALPs, the paper considers a prompt diphoton bump hunt with EPA-based estimates of light-by-light, pi0-pair, and omega backgrounds and S/sqrt(B) = 2, and a displaced diphoton search treated as background-free with 3 signal events. The claimed reach covers dark photon masses from about 2 me to 3 GeV and couplings epsilon of order 1e-6 to 1e-3, and ALP masses 0.1 to 5 GeV with 1/Lambda down to about 1e-6 GeV^-1.","tokens_in":15160,"tokens_out":11563,"duration_ms":114384,"significance":"If validated, this would be a useful and timely set of projections for a proposed facility and would strengthen the case for a BSM physics program at the EicC. The production cross-section formulas are standard and explicitly documented in Appendix A, and the prompt ALP analysis makes a concrete attempt to model the dominant backgrounds, which is a genuine strength. The paper also credits previous EIC work and clearly separates the two detector-length scenarios. The main quantitative claims, however, rest on two assumptions that are not yet supported: background-free (or unquantified-background) displaced searches and detector performance parameters taken from an unpublished CDR. For these reasons the paper is a promising feasibility study rather than a demonstrated sensitivity projection.","major_comments":[{"comment":"Equations (2.6)-(2.7) define the displaced dark-photon reach by requiring N_A' = 10 or 100 signal events, but no SM background estimate is given. In coherent e-Pb collisions at 10 fb^-1, pi0/eta photoproduction with Dalitz decays and photon conversions inside the 14 cm to 1 m decay volume can produce e+e- pairs with non-prompt vertices, and the paper specifies no invariant-mass window, vertex-quality cut, or conversion veto. The statement that the N_A' = 100 bound would still cover unexplored parameter space even with backgrounds is not a substitute for a background estimate, and it does not justify the N_A' = 10 curve that sets the lower edge of the claimed reach. Please provide a background estimate with the relevant rejection cuts, or relabel the curves as raw signal-yield contours.","section":"Section 2.1, Fig. 5"},{"comment":"The signal yield for mA' above the two-pion threshold is not defined carefully. The text says only lepton pairs are used as the signal and quotes Eq. (2.5) for the leptonic partial width, but the sensitivity curves in Fig. 5 extend to about 3 GeV, where hadronic decay modes dominate the total width and reduce Br(A' -> l+l-). The manuscript never states whether N_A' includes this branching ratio and uses the total width in Eq. (2.7). If Eq. (2.5) is used as a total width, the high-mass lifetimes are overestimated and the reach is optimistic. Please write the explicit formula for N_A' in terms of the production cross section, branching ratio, detection efficiencies, and the decay probability, and recompute Fig. 5 accordingly.","section":"Section 2.1, Eq. (2.5), Fig. 5"},{"comment":"The displaced ALP search is declared background-free and normalized to N_a = 3, with no estimate of how many e-Pb events produce a reconstructed diphoton vertex in the 0.1 to 1 m window. Prompt pi0/omega backgrounds are not shown to be rejected by the vertex resolution or by the cuts in Eqs. (3.4)-(3.7), and photon conversions inside the decay volume are not discussed. In addition, the single-photon detection efficiency is assumed to be the same as the lepton efficiency f(E,eta) about 0.9 from Fig. 4 without an independent justification for photons. Without these inputs, Fig. 11 cannot support the statement that the EicC has enhanced sensitivity for long-lived ALPs compared with upcoming experiments.","section":"Section 3.2, Fig. 11"},{"comment":"The central numerical results depend on detector parameters from the unpublished EicC CDR, Ref. [41]: L1 = 10 micrometres, L2 = 0.14 m or 1 m, LR = 0.1 m, LEM = 1 m, the resolution parameters in Table 2, and the efficiency map f(E,eta) about 0.9. The paper gives no variation of these parameters, so the reader cannot tell how much the reach would shrink if, for example, the vertex resolution or decay volume differ from the CDR assumptions. Please add a scan over the most critical parameters (at least L1 and L2 for dark photons, and LR and LEM for ALPs) and state explicitly that the projections are contingent on these values.","section":"Sections 2.1 and 3.2, Appendix B"},{"comment":"The prompt ALP limits are derived from S/sqrt(B) = 2 with background cross sections estimated via the EPA for light-by-light, pi0-pair, and omega production, but no theoretical uncertainties are assigned to these estimates. The pi0-pair rate uses the handbag model in the high-mass region and the omega rate relies on 1973 photoproduction data; an O(1) uncertainty in B translates into a tens-of-per-cent shift in the 1/Lambda boundary. Please provide an uncertainty band or at least a concise discussion of the model dependence of the quoted reach.","section":"Section 3.1, Fig. 10"}],"minor_comments":[{"comment":"Figure 4 contains a Mathematica output artifact, 'Out[ ] = 0.', in the left panel; it should be removed.","section":"Figure 4"},{"comment":"The Figure 11 caption contains a typo, 'banchmark luminosity', which should read 'benchmark luminosity'.","section":"Figure 11"},{"comment":"Equation (3.2) is written as a proportionality with ambiguous placement of the factors in the denominator; since the normalization is essential for the sensitivity, please define t_e and t_N and give the explicit expression or a precise reference.","section":"Eq. (3.2)"},{"comment":"Reference [50] is incomplete: it lacks a publication date and an arXiv identifier; please update it.","section":"References"},{"comment":"In Table 2, the entry c2 = -1.5 x 10^-6 at eta = 2.7 has a qualitatively different magnitude from the other entries; please confirm that this is not a typographical error.","section":"Table 2"}],"recommendation":"major_revision","confidential_remarks":"The paper would benefit from coordinating with the EicC collaboration on the citation of the unpublished CDR, Ref. [41], since several reach curves depend on numbers from that document. The self-citation to Ref. [36] is acceptable because that paper is published and the ALP matrix element is standard."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The useful thing here is a set of EicC sensitivity projections for dark photons and ALPs, based on standard coherent-scattering calculations adapted to a lower beam energy. The lower-boost point is real: it makes decay products more central and improves displaced-vertex detection relative to the EIC. The paper gives credit where it is due: cross-section formulas are standard and referenced, the ALP prompt search includes explicit background estimates (LBL, pi0 pairs, omega) and uses S/sqrt(B), and the appendices provide enough detail to reconstruct the calculation. The resolution parameters table is a useful compilation.\n\nThe main soft spot is the dark photon displaced search. In Section 2.1, sensitivity is defined by requiring N_A'=10 or 100 signal events with no SM background model. The text acknowledges this, and the stress-test note correctly identifies that coherent photoproduction of pi0/eta, Dalitz decays, and photon conversions can produce e+e- pairs with displaced vertices in the decay volume. No invariant-mass window, vertex-quality cut, or conversion veto is specified. So the N=10 curve, which sets the lower edge of the claimed dark-photon reach, is not yet justified. This is significant but not fatal: the N=100 curve may still cover large areas, and the paper states that. Still, the abstract's \"comprehensively probe previously unexplored parameter space\" overstates what is demonstrated.\n\nTwo other caveats are proportionate. The detector parameters (vertex resolution, decay-volume sizes, efficiencies) come from an unpublished CDR, and no code or data is provided. That makes the projections hard to check, though not inherently wrong. The self-citation to Ref. [36] is not a problem: the matrix element is published, and shared authorship does not create circularity here.\n\nOverall, the paper is a competent physics-case study for a proposed machine, not a measurement or a new mechanism. The math and standard references look solid. The main issue is the background assumption in the displaced dark-photon search. This paper deserves a serious referee; a good referee should ask the authors to add a background estimate or defend the background-free assumption, and to make the detector parameters publicly available. I would bring it to a reading group focused on long-lived particle searches or EicC physics, and I would cite it as a projection for those channels.","headline":"A useful sensitivity projection for dark photons and ALPs at EicC, worth engaging, but the displaced dark-photon reach needs a background model before the gap-filling claim is fully supported.","tokens_in":15657,"tokens_out":2213,"would_cite":true,"duration_ms":23577,"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 projects that the Electron-Ion Collider in China could detect dark photons with masses from $2\\,m_e$ to 3 GeV and couplings down to $10^{-6}$, and axion-like particles from 0.1 to 5 GeV with couplings down to…","keywords":["dark photon","axion-like particle","Electron-Ion Collider in China","coherent scattering","displaced vertex search","bremsstrahlung","photon fusion","beyond Standard Model"],"falsifier":"Measure the actual EicC vertex resolution, decay volume, and photon/lepton efficiency once the detector exists; for example, if the charged-vertex resolution is worse than 10 um or the electromagnetic calorimeter is farther than 1 m, the dark photon and ALP displaced-vertex sensitivities in the paper's figures move to larger couplings and may no longer exceed current bounds.","tokens_in":14789,"feed_emoji":"⚛️","tokens_out":9155,"duration_ms":83486,"temperature":0.7,"pith_summary":"This paper projects the sensitivity of the proposed Electron-Ion Collider in China (EicC) to two kinds of feebly interacting particles: dark photons and axion-like particles (ALPs) coupled to photons. Its central claim is that the EicC's moderate 3.5 GeV electron beam is an asset: particles produced by coherent scattering off lead nuclei are less boosted than at higher-energy electron-ion colliders, so their decay products fall inside the detector acceptance, and displaced-vertex searches can cover dark photon masses from $2\\,m_e$ to 3 GeV with kinetic-mixing couplings $\\epsilon$ from $10^{-6}$ to $10^{-3}$. For ALPs, the paper argues that prompt diphoton and displaced-vertex searches together reach masses from 0.1 to 5 GeV and effective couplings $1/\\Lambda$ down to about $10^{-6}\\,\\mathrm{GeV}^{-1}$, exceeding current bounds. If the projections hold, a relatively modest electron-ion collider could map a large slice of unexplored new-physics parameter space before or alongside higher-energy machines.","feed_headline":"EicC could probe dark photons and axions in untouched gaps","feed_subtitle":"A proposed Chinese electron-ion collider could map dark photon and axion signals today's experiments miss.","key_machinery":"The argument is carried by two production mechanisms and one detection signature. Dark photons are produced via electron bremsstrahlung on a lead nucleus, and ALPs via photon fusion, both in the coherent regime where the momentum transfer to the nucleus is smaller than the inverse nuclear radius, giving a $Z^2$ enhancement and a nuclear form-factor suppression at larger transfer. The signal is a displaced vertex: for dark photons, a reconstructed lepton pair with decay probability $e^{-L_1/L_{A'}} - e^{-L_2/L_{A'}}$ between the vertex resolution $L_1\\simeq 10\\,\\mu\\mathrm{m}$ and the detector length scale $L_2$; for ALPs, a diphoton vertex between $L_R = 0.1$ m and the electromagnetic calorimeter distance $L_{\\mathrm{EM}} = 1$ m. The efficiency map $f(E,\\eta)\\simeq 0.9$ from the EicC detector concept converts these geometric acceptances into event counts.","core_discovery":"On the paper's own terms, the discovery is that the EicC is not simply a lower-energy copy of a larger electron-ion collider but a complementary probe: its lower electron-beam energy gives the produced dark photons and ALPs a smaller boost, making their displaced decay vertices detectable with a vertex resolution $L_1\\simeq 10\\,\\mu\\mathrm{m}$ and a decay volume of $L_2 = 1$ m (or 0.14 m for the inner barrel). The production rates come from $Z^2$-enhanced coherent scattering, with the dark photon produced by electron bremsstrahlung and the ALP by photon fusion; the paper computes cross sections using the nuclear elastic form factor and requires only 10--100 dark-photon events or $S/\\sqrt{B}=2$ for the ALP prompt search. The result is a projected exclusion or discovery reach that spans the gap between meson-decay and beam-dump constraints for dark photons and exceeds current bounds for ALPs in the 0.1--5 GeV mass window.","pith_inferences":["The same lower-boost argument extends to other light weakly coupled states, such as heavy neutral leptons or millicharged particles, where displaced-vertex detectability would also improve at the EicC.","If the real detector achieves worse than the quoted 10 $\\mu$m vertex resolution or lower than 0.9 single-particle efficiency, all sensitivity curves shift to larger couplings; measuring those parameters is the fastest test of the paper's central claim.","The complementarity argument suggests an ordering principle: for displaced-vertex searches, lower beam energy trades production rate for detectability, so the optimal energy depends on the target coupling window, and the EicC's 3.5 GeV is tuned to the $10^{-6}$ to $10^{-3}$ coupling range.","The ALP prompt-search reach could be checked with a fast simulation of the actual calorimeter resolution before the collider is built, since the invariant-mass resolution table drives how strongly the light-by-light, pion-pair, and omega backgrounds are suppressed."],"forward_implications":["A 10 fb$^{-1}$ run with $L_2=1$ m would exclude or discover dark photon couplings down to $\\epsilon\\sim10^{-6}$ across $2\\,m_e \\lesssim m_{A'} \\lesssim 3$ GeV, closing the gap between meson-decay and beam-dump limits.","The same machine would reach ALP couplings $1/\\Lambda$ down to roughly $2\\times10^{-5}$ GeV$^{-1}$ in the prompt search at 100 fb$^{-1}$ for $0.1 \\lesssim m_a \\lesssim 5$ GeV, surpassing current bounds in that mass window.","Because the EicC's lower boost lengthens lab-frame lifetimes, its displaced-vertex ALP search is more sensitive to long-lived ALPs than a higher-energy electron-ion collider search with the same detector concept.","A future upgrade to 100 fb$^{-1}$ and a required signal of 3 events would push the dark photon reach to smaller couplings, covering more of the so-far unexplored region.","The projected sensitivities complement, rather than duplicate, planned beam-dump and high-energy collider searches, since each experiment covers a different coupling-lifetime slice."],"supporting_citations":[{"why":"Supplies the ALP production cross-section calculation, matrix element, and light-by-light background treatment that this paper adapts to the EicC.","marker":"[36]"},{"why":"Defines the EicC design, beam energies, and detector concept used for all sensitivity estimates.","marker":"[40, 41]"},{"why":"Provides the vertex resolution L1 = 10 um, decay lengths L2 = 1 m / 0.14 m, LR = 0.1 m, LEM = 1 m, the efficiency f(E, eta) ~ 0.9, and the Table 2 resolution parameters.","marker":"[41]"},{"why":"Gives the dark photon decay width to lepton pairs used in the displaced-vertex probability.","marker":"[43]"},{"why":"Provides the elastic nuclear form factor that suppresses coherent production at large momentum transfer.","marker":"[85]"},{"why":"Supplies the amplitude calculation for coherent production of a light new vector or pseudoscalar used in the dark photon cross section.","marker":"[86]"},{"why":"Establishes the equivalent photon approximation used for both ALP production and the light-by-light, pion-pair, and omega backgrounds.","marker":"[65, 66]"},{"why":"Provides the QED corrections to light-by-light scattering used for the irreducible ALP search background.","marker":"[67]"}],"fun_headline_variants":["EicC to fill dark photon and axion blind spots","EicC could hunt dark photons and axions in gaps","China's EicC to probe dark photon and axion gaps","EicC may reveal dark photons and axions beyond current bounds","EicC to probe dark photons and axions in new mass windows"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The projections assume the EicC detector will achieve the vertex resolution (L1 ~ 10 um), decay volumes (L2 = 1 m or 0.14 m; LR = 0.1 m; LEM = 1 m), and average single-particle efficiency f(E, eta) ~ 0.9 quoted from the unpublished EicC Conceptual Design Report; if the real detector performs worse, the reach shrinks and the claimed gap-filling coverage disappears.","fun_headline_variants_meta":{"raw":{"variants":["EicC to fill dark photon and axion blind spots","EicC could hunt dark photons and axions in gaps","China's EicC to probe dark photon and axion gaps","EicC may reveal dark photons and axions beyond current bounds","EicC to probe dark photons and axions in new mass windows"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00116,"raw_usage":{"total_tokens":4800,"prompt_tokens":937,"completion_tokens":3863,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":553,"completion_tokens_details":{"reasoning_tokens":3773}},"tokens_in":553,"tokens_out":3863,"duration_ms":25440,"temperature":1.0,"reasoning_tokens":3773,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T19:48:13.775218+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the actual EicC vertex resolution, decay volume, and photon/lepton efficiency once the detector exists; for example, if the charged-vertex resolution is worse than 10 um or the electromagnetic calorimeter is farther than 1 m, the dark photon and ALP displaced-vertex sensitivities in the paper's figures move to larger couplings and may no longer exceed current bounds.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the vertex resolution L1 = 10 um, decay lengths L2 = 1 m / 0.14 m, LR = 0.1 m, LEM = 1 m, the efficiency f(E, eta) ~ 0.9, and the Table 2 resolution parameters."}],"review_version":1}