{"id":"e61e0a8c-cb7b-48d4-969a-bf08fd122355","arxiv_id":"2510.23065","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"A cut-based analysis of e+e−→ννγ at a Tera-Z factory is projected to constrain BR(Z→ννγ) to about 2×10^-9 (3σ, no systematics), an order-3 improvement over LEP but above the SM value 7.2×10^-10.","lead":"A simulation study estimates how well future Tera-Z runs of FCC-ee/CEPC could constrain the rare decay Z→ννγ and its anomalous couplings. It projects branching-ratio limits near 10^-9, roughly three orders of magnitude better than LEP, though still above the Standard Model prediction.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Missing SM-EFT interference in e+e-→ννγ can shift projected BR limits by an order of magnitude; a full-model simulation check is needed before the Tera-Z sensitivity claim is used.","rationale":"I read the paper as a standard cut-based projection, and the claimed improvement of several orders of magnitude over LEP is plausible if all amplitudes are included. The most load-bearing step is the separate generation of signal and SM ννγ background, followed by incoherent addition in the significance formula. The identical final state means interference is not optional; it is part of the SM+EFT amplitude. The size of the interference is not estimated anywhere in the manuscript. A quick numerical check with the same UFO/MadGraph framework would settle the magnitude. Because the interference could either enhance or suppress the reach, the correct verdict is conditional: accept only after this check is done. I do not see this as a reason to reject outright; the simulation framework is standard, and the authors have provided enough detail to reproduce the test. The reader's weakest-assumption point matches mine, so I agree with the reader's identification. I also note the table-normalization issue (Table I event counts seem inconsistent with the analytic BR for κ=5 TeV⁻²), but the interference check is the single decisive test that would validate or invalidate the central claim.","tokens_in":12246,"tokens_out":19517,"duration_ms":214763,"concrete_test":"Generate e+e−→ννγ in the full EFT+SM model (all diagrams) at a benchmark κ/Λ²=5 TeV⁻² and α8/Λ⁴=15 TeV⁻⁴ with the same cuts as Table I, and compare the post-S_ET>16 yield with the sum of separately generated 'signal-only' (Fig. 4a) and SM-background samples. Run both signs of κ and α8. If the full-model yield differs from the incoherent sum by more than the statistical precision, recompute the 3σ/5σ BR limits including interference; if those limits shift by more than a factor of 2, the central claim requires revision.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central projection treats the EFT signal (Fig. 4a) and the SM e+e−→ννγ background (Fig. 4b–d) as incoherent samples combined through S+B in Eq. (9). But both amplitudes have identical external states, so the physical rate is |M_SM + M_EFT|². The interference term 2Re(M_SM M_EFT*) can easily dominate the pure |M_EFT|² contribution: after the S_ET>16 cut, the signal yield for a BR near 10⁻⁹ is roughly 10⁻³ of the total background, so even a modest coherent phase yields an interference term ∼2√(SB) ∼ 0.06B, i.e. tens of times the pure signal. No argument is given for why the interference should vanish (helicity, kinematics, or CP), and the paper does not quantify the effect on Table II. Since the limits on κ/Λ² and α8/Λ⁴, and hence the inferred BR limits, are obtained from the significance curves that use this S, the neglect of interference is the least secure assumption in the chain. The issue is not an internal inconsistency in the simulation pipeline, but a missing contribution to the same final state that is in principle calculable with the same MadGraph setup.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a search for anomalous Zννγ couplings through e+e−→ννγ production at the FCC-ee/CEPC Tera-Z stage (√s=91.2 GeV, L_int=150 ab^-1). The authors implement dimension-6 (κ1,κ2/Λ^2) and dimension-8 (α8/Λ^4) operators in a UFO model, generate signal and SM background events with MadGraph/Pythia/Delphes, and apply a cut-based selection using Eγ, /ET, and S_ET. From significance curves they derive projected upper limits on the couplings and, via the analytic width relations of Eqs. (7)–(8), on BR(Z→ννγ), obtaining values of order 10^-9 at δ_sys=0 and a few times 10^-7 at δ_sys=5% (Table II). The paper claims this is an improvement of several orders of magnitude over the LEP bound and demonstrates the potential of Tera-Z to test rare Z decays.","tokens_in":12632,"tokens_out":6137,"duration_ms":65465,"significance":"If the projected sensitivities are correct, the analysis would be a useful addition to the Tera-Z physics case: the Z→ννγ channel is not strongly constrained by LEP, and a future Z-pole run could test loop-level SM predictions. The use of a full simulation chain (MadGraph+Pythia+Delphes) with a detailed cut-flow table is a strength, and the analytic decay-width expressions provide a transparent mapping between couplings and branching ratios. However, the central quantitative claims are not yet supported because two key ingredients are missing: (i) interference between the EFT amplitude and the SM e+e−→ννγ amplitude, which is not modeled, and (ii) an explicit charged-lepton veto, despite a large surviving l+l−γ background in Table I. Both are standard, calculable additions and could change the limits in Fig. 9 and Table II substantially. The paper's broad conclusion is plausible, but the specific numbers should be treated as preliminary until these omissions are addressed.","major_comments":[{"comment":"The signal (Fig. 4a) and the SM e+e−→ννγ background (Fig. 4b–d) are generated separately and combined as S+B in Eq. (9), but both amplitudes have identical external states. The physical rate is |M_SM + M_EFT|^2, and the interference term 2Re(M_SM M_EFT*) is absent from the analysis. No helicity, CP, or kinematic argument is given for its vanishing; it is in principle computable in the same MadGraph setup. Since the coupling limits in Fig. 9 and the BR limits in Table II are extracted from these S values, this omission is load-bearing. The authors should generate the complete e+e−→ννγ process including both SM and EFT contributions, or present an explicit amplitude-level argument showing the interference vanishes after integration over the selected phase space.","section":"Sec. II, Eq. (9)"},{"comment":"The cut-flow table reports 5.37×10^6 surviving l+l−γ events after the final S_ET>16 requirement, comparable to the 8.17×10^6 SM ννγ background. The text never describes a charged-lepton veto or isolation requirement; the baseline selection is only Nγ>0. Since l+l−γ events contain two visible charged leptons, they cannot mimic a single-photon + missing-energy signature unless both leptons are lost. A standard lepton veto would suppress most of this background. As written, the background composition, significance curves (Fig. 9), and resulting Table II limits are not representative of a genuine ννγ selection. A documented lepton veto must be applied, or the survival mechanism for these events must be demonstrated.","section":"Table I, Sec. II"}],"minor_comments":[{"comment":"Several entries appear to have typographical issues: the α8/Λ4 3% δsys, 5σ entry reads '3.72×10^7' and should presumably be 10^-7; also the α8/Λ4 5% δsys 3σ and 5σ limits (6.18×10^-7 and 6.19×10^-7) are almost identical, which is suspicious since a 5σ limit should be weaker than a 3σ limit. Please check and correct.","section":"Table II"},{"comment":"The integration limits for the decay width are described as 'minimum value for the photon energy greater than half of the beam energy' and maximum equal to M_Z. The relation between these limits and the analysis cuts (Eγ>4 GeV) is not transparent; clarify whether the width used for Figure 3 and Table II matches the phase space accepted by the experimental selection.","section":"Sec. I, Eqs. (7)–(8)"},{"comment":"The left-panel caption states κ/Λ^2=5.0 TeV^-4, while the text and other figures use TeV^-2; this is presumably a typo. Please correct.","section":"Fig. 7"},{"comment":"The text names e+e−→γγγ as a subleading background, but Table I does not include it in the cut-flow. Either include it in the table or explain why it is negligible after the selection.","section":"Sec. II, Table I"},{"comment":"The claim of 'several orders of magnitude' improvement over LEP is only valid for δ_sys=0. At δ_sys=5%, the limits in Table II are a few times 10^-7, which is only a factor of about 3 below the LEP bound of 10^-6. Please qualify this statement in the abstract and conclusion.","section":"Abstract and conclusion"},{"comment":"The limits on BR(Z→ννγ) in Table II are model-dependent translations of coupling limits via Eqs. (7)–(8), not direct, model-independent measurements. The text should state this more explicitly to avoid the impression that the analysis constrains the SM branching ratio itself.","section":"Sec. III"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a standard, readable Tera-Z projection for Z→ννγ using dimension-6 and dimension-8 EFT operators. The authors implement the operators in FeynRules/MadGraph, run a full simulation with Pythia/Delphes IDEA, and derive expected limits on BR(Z→ννγ) from a cut-based analysis. The central message—an improvement of two to three orders of magnitude over the LEP bound—is plausible and worth having.\n\nWhat is new: the specific simulation for FCC-ee/CEPC with the IDEA detector card, the simultaneous treatment of d=6 and d=8 operators, and the cut-flow table. The analysis pipeline is normal for the genre, and the authors are transparent about the cuts.\n\nThe soft spots, in order of importance:\n\n1. The interference between the EFT amplitude and the SM ννγ background is not included. Both amplitudes produce the same final state, so the physical rate is |M_SM+M_EFT|². The paper generates signal and background separately and combines them incoherently as S+B. For the couplings that saturate the quoted 3σ limits (BR ~1e-9), the signal after cuts is a small fraction of the background, so the interference term 2Re(M_SM M_EFT*) can be comparable to or larger than the pure EFT term. The paper gives no helicity/kinematics argument for why this term should vanish, and no estimate of its effect on Table II. This is the main quantitative gap. It is fixable—the interference can be computed with the same MadGraph setup by generating the full process with both SM and EFT contributions in one matrix element.\n\n2. The background treatment is underspecified. There is no explicit charged-lepton veto, yet the l+l-γ background survives at the same level as the signal after the final cut. The cuts listed (Eγ>4 GeV, /ET>4 GeV, /ET≤Eγ, S_ET>16) reduce l+l-γ by four orders, but a lepton veto is standard and should be documented. If it is implicitly included in the detector simulation, say so.\n\n3. Typos and overstatements. Table II has a missing minus sign for α8 at δsys=3% (3.72×10^7 should be 3.72×10^-7). Fig. 7 writes TeV^-4 instead of TeV^-2 for κ/Λ². The abstract says the SM prediction is \"four times smaller\" than the LEP limit; it is about 1400 times smaller. And the claim that the projected limits will \"test the SM loop-level prediction\" is not supported: the best 3σ limit (1.73e-9) is still a factor of 2.4 above the SM prediction (7.16e-10), so the analysis can at best set an upper limit, not test the SM value.\n\nWho it is for: people doing Tera-Z rare-decay studies will find the simulation setup and the cut-based workflow useful. But the numeric limits should not be used until the interference is quantified.\n\nRecommendation: send to peer review, but with the clear expectation that the authors address the interference term and correct the typos. The qualitative conclusion is likely stable, but the headline numbers are not yet trustworthy.","headline":"Useful Tera-Z projection for Z→ννγ, but the missing SM-EFT interference term and some typos make the headline BR limits less secure than they look.","tokens_in":13097,"tokens_out":7813,"would_cite":false,"duration_ms":85620,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper projects that a Tera-Z electron-positron collider can constrain the rare decay Z→ννγ to branching ratios near 10^-9, several orders of magnitude below the current LEP bound.","keywords":["Z boson rare decays","Tera-Z factory","monophoton signature","effective field theory","anomalous Zννγ couplings","neutrino-photon interaction","future e+e− collider","branching ratio limits"],"falsifier":"Compute, for a benchmark such as κ/Λ^2=5 TeV^-2 or α8/Λ^4=15 TeV^-4, the complete e+e−→ννγ matrix element containing both the SM and EFT amplitudes, apply the same selection (Eγ>4 GeV, /ET>4 GeV, /ET≤Eγ, S_ET>16), and compare the event count with the paper's independent S+B sum. A difference comparable to the statistical uncertainty would mean the quoted limits need revision; a difference much smaller would support them.","tokens_in":12155,"feed_emoji":"⚛️","tokens_out":8330,"duration_ms":69499,"temperature":0.7,"pith_summary":"The paper argues that a future Tera-Z electron-positron collider—running at the Z pole with 150 inverse attobarns—can use the single-photon production process e+e−→ννγ to measure or constrain the rare decay Z→ννγ far beyond present bounds. The Standard Model predicts this decay at one loop with branching ratio 7.16×10^-10, four times below the existing LEP limit near 10^-6. By parameterizing anomalous Zννγ couplings through dimension-6 and dimension-8 effective operators, simulating signal and background events, and applying a simple cut sequence on photon energy, missing transverse energy, and missing-energy significance, the authors obtain projected upper limits on BR(Z→ννγ) in the 10^-9 range under ideal systematics. Even with 5% systematic uncertainty the projected limits remain orders of magnitude below LEP. If correct, this makes the Tera-Z program a direct test of the loop-level Standard Model and a sensitive search for new physics in neutrino-photon interactions.","feed_headline":"Paper projects Tera-Z limit on Z→ννγ near 10^-9","feed_subtitle":"The rare decay's SM rate is seven parts per ten billion, and a Tera-Z run could be the first to test it.","key_machinery":"The central object is an effective Lagrangian for the Zννγ vertex: a dimension-6 term with CP-odd and CP-even couplings κ1/Λ^2 and κ2/Λ^2, and a dimension-8 term with coupling α8/Λ^4, added to the Standard Model. The authors implement this vertex in a Monte Carlo event generator with detector simulation, then use three kinematic handles—the photon energy Eγ, the missing transverse energy /ET, and its significance S_ET=/ET^2/E_γ^T—to separate the monophoton signal from the dominant e+e−→ννγ and e+e−→ℓ+ℓ−γ backgrounds. The cut sequence and the significance formula in Eq. (9) carry the whole projection: they convert the difference between signal-plus-background and background alone into the rep","core_discovery":"On its own terms, the paper establishes a sensitivity projection: with √s=91.2 GeV and L_int=150 ab^-1, the e+e−→ννγ channel, after cuts requiring a photon with Eγ>4 GeV, missing transverse energy /ET>4 GeV with /ET≤Eγ, and missing-energy significance S_ET>16, yields an expected 3σ upper limit on BR(Z→ννγ) of 1.73×10^-9 for the dimension-6 coupling κ/Λ^2 (and 1.77×10^-9 for the dimension-8 coupling α8/Λ^4) when systematic uncertainty is neglected. The same analysis reaches 5σ evidence for κ/Λ^2≲1 TeV^-2 and α8/Λ^4 on the order of 10 TeV^-4. With 5% systematics the limits degrade to about 10^-7, still well below the LEP bound. The paper treats the EFT signal simply as an additive contribution","pith_inferences":["Not addressed in the paper: the EFT signal and SM ννγ background are generated separately and summed (S+B) without computing the interference between the new-physics vertex and the SM amplitudes, even though they share the same final state. If that interference is non-negligible, the quoted κ/Λ^2 and α8/Λ^4 limits—and the BR values in Table II—could shift appreciably.","A testable extension is to run the same analysis at the off-peak energies shown in the paper (87.7 and 93.9 GeV): the ratio of event rates would separate the resonant Z-pole signal from non-resonant backgrounds and directly check the EFT signal hypothesis.","A further extension: because the dimension-6 Lagrangian contains both a CP-odd (κ1) and a CP-even (κ2) coupling, a photon angular or polarization analysis could distinguish these two couplings rather than only constraining the sum κ1^2+κ2^2, which the current integrated rate cannot do."],"forward_implications":["A Tera-Z run can test the one-loop Standard Model prediction for Z→ννγ (7.16×10^-10) for the first time, since the projected 3σ reach at δ_sys=0 is about 1.7×10^-9, within a factor of roughly 2.4 of the SM value.","New-physics couplings κ/Λ^2 and α8/Λ^4 can be bounded at the TeV^-2 / TeV^-4 scale, probing new physics scales far beyond direct reach.","A simple cut-based monophoton selection is sufficient to suppress the huge e+e−→ℓ+ℓ−γ background by roughly four orders of magnitude while keeping about 5% of the signal.","Even at 5% systematic uncertainty, the projected limits (≈3.6×10^-7) improve on the LEP bound of ≈10^-6, so the Tera-Z program has discovery potential even in a conservative scenario."],"fun_headline_variants":["Tera-Z projects Z→ννγ limit at 1.7×10^-9","Rare Z decay testable at Tera-Z: limit near 10^-9","New physics via Z→ννγ probed to 10^-9 at Tera-Z","Tera-Z could first test SM's Z→ννγ loop rate","Projected limit on Z→ννγ from Tera-Z: 10^-9"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The load-bearing premise is that the anomalous Zννγ signal and the Standard Model ννγ background can be added independently, with no interference between the EFT coupling and the SM one-loop amplitude; if that interference is significant, the projected branching-ratio limits change.","fun_headline_variants_meta":{"raw":{"variants":["Tera-Z projects Z→ννγ limit at 1.7×10^-9","Rare Z decay testable at Tera-Z: limit near 10^-9","New physics via Z→ννγ probed to 10^-9 at Tera-Z","Tera-Z could first test SM's Z→ννγ loop rate","Projected limit on Z→ννγ from Tera-Z: 10^-9"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000899,"raw_usage":{"total_tokens":3795,"prompt_tokens":921,"completion_tokens":2874,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":665,"completion_tokens_details":{"reasoning_tokens":2760}},"tokens_in":665,"tokens_out":2874,"duration_ms":19039,"temperature":1.0,"reasoning_tokens":2760,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-04T07:59:26.902901+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute, for a benchmark such as κ/Λ^2=5 TeV^-2 or α8/Λ^4=15 TeV^-4, the complete e+e−→ννγ matrix element containing both the SM and EFT amplitudes, apply the same selection (Eγ>4 GeV, /ET>4 GeV, /ET≤Eγ, S_ET>16), and compare the event count with the paper's independent S+B sum. A difference comparable to the statistical uncertainty would mean the quoted limits need revision; a difference much smaller would support them.","supporting_citations":[],"review_version":1}