{"id":"de014112-9ce6-45a5-aa3a-5df156aa46ba","arxiv_id":"2412.10315","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":7,"one_line_summary":"A 10 kg forward neutrino detector at a 3 TeV muon collider could probe neutrinophilic scalar mediators down to couplings that reach thermal freeze-out and freeze-in dark matter relic targets.","lead":"A future muon collider would produce an intense, well-understood neutrino beam; the paper shows a small detector in that beam could search for a new particle that carries dark matter's interactions with neutrinos. Because even a 10 kg detector could beat other proposed experiments, the result gives a concrete physics case for a forward neutrino detector at the muon collider.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Charge misidentification of muons is the load-bearing unquantified assumption: the claimed background rejection rests on vetoing all muons above 30 GeV by sign, but no mis-ID rate is modeled, and even a rate near 1e-3 could inject many CC fakes before the BDT.","rationale":"I read the paper as a carefully executed sensitivity forecast rather than a measurement, and I find no fatal internal inconsistency. The signal kinematics, the pure nu_mu forward beam, and the wrong-sign muon signature are well matched; the cross-section normalization is cross-checked against semi-analytic estimates from Ref. [18]; and the cut table is transparent enough to expose where the assumptions enter. The single most load-bearing point is the perfect muon-sign identification assumption flagged by the reader and by the paper itself. The first two cuts use charge sign to discard roughly 5e7 backgrounds, so any nonzero mu- to mu+ misidentification contaminates the signal region with CC events. The expected rate scales as epsilon times the CC event count, and the paper provides no estimate of epsilon from momentum resolution, no fake-sample study, and no BDT validation on a mis-ID-inclusive sample. This is not a disagreement with consensus physics; it is a quantitative detector-performance assumption that the sensitivity claim depends on. Because the reader already assigned CONDITIONAL acceptance, my read does not move the verdict: the paper should be accepted conditional on either a quantitative muon charge-mis-ID study, or release of the simulation and flux files so that the effect can be evaluated independently. The concrete test above would settle whether the concern actually degrades the reach.","tokens_in":17477,"tokens_out":10465,"duration_ms":106743,"concrete_test":"Re-run the full 10 kg, one-year analysis with a charge-misidentification model: for each generated CC nu_mu event containing a reconstructed mu- with E>30 GeV, flip its charge with probability epsilon in {1e-4, 1e-3, 1e-2}, then apply the same preselection, kinematic cuts, and BDT (retraining the BDT on the mixed sample if needed). Report the final number of mis-ID-induced signal-like events and the shift in the m_phi=1 GeV exclusion curve. If epsilon=1e-3 alone yields more than roughly one final fake event, or a greater-than-factor-two shift in lambda, the perfect-sign assumption is load-bearing; the same exercise should be repeated for the 10 ton detector, where the final background is already about 100 events.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is a 10 kg-yr sensitivity forecast in which the signal is a wrong-sign muon, nu_mu N -> mu+ phi X. The analysis suppresses about five orders of magnitude of background by requiring one positive muon above 100 GeV and no negative muon above 30 GeV (Table I); only 6.93e3 events survive these two cuts, and 2.23 survive the full BDT chain. The paper states in Sec. III, under 'Positive muon': 'For simplicity, we assume perfect muon sign identification within this setup and above the energy threshold. It is important to emphasize that muon sign identification is crucial for the analysis presented below.' No charge-misidentification rate is assigned. In a real magnetized spectrometer, a true mu- from a CC nu_mu event that is reconstructed as mu+ can pass the baseline selection: there is no true mu+, so the mu- veto is not triggered, and the reconstructed mu+ energy is typically high. With roughly 1.89e7 CC nu_mu events per year in the 10 kg detector, even epsilon=1e-3 yields ~1.9e4 such misidentified candidates before kinematic cuts, and epsilon=1e-4 yields ~1.9e3, comparable to the entire baseline background. The manuscript does not show that the Evis, pT, Delta-phi, charm-tag, and BDT selections remove this contamination, nor is the BDT trained or evaluated on a mis-ID-inclusive sample. Since the quoted lambda bounds and the claimed reach to thermal freeze-out and freeze-in targets depend on the final background of 2.23 events, this is a load-bearing detector-performance assumption rather than an internal theory inconsistency.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper proposes a forward neutrino detector (MuColν) at a 3 TeV muon collider and studies its sensitivity to a neutrinophilic scalar φ that couples to muon neutrinos through the dimension-six operator in Eq. (1) and interacts with dark matter. The signal is the mono-neutrino process ν_μ N → μ^+ φ X, identified by requiring a high-energy positive muon and no high-energy negative muon. The authors simulate signal and Standard Model backgrounds with MadGraph/Pythia, validate the signal cross section against the semi-analytic estimate in Eq. (4), and present both a cut-based analysis and a Boosted Decision Tree analysis, with event counts summarized in Table I. They derive projected exclusion bounds for 10 kg and 10 ton detectors, compare them with DUNE, FLArE, IceCube-Gen2, and Higgs-invisible-width searches, and argue that even a 10 kg-yr exposure can probe thermal freeze-out (Dirac fermion and complex scalar) and sterile-neutrino freeze-in targets for m_φ from roughly 1 MeV to 20 GeV.","tokens_in":17920,"tokens_out":7307,"duration_ms":667770,"significance":"If the projection holds, this is a valuable and timely physics case for a forward neutrino detector at a muon collider: the muon-decay neutrino flux is theoretically very well understood, the wrong-sign-muon signature is a clean and motivated probe of lepton-number-carrying mediators, and the projected reach in λ_μμ would exceed other terrestrial projections in the few-MeV to 10-GeV window. The paper has several concrete strengths: the signal cross sections are checked against Eq. (4), the background rejection is presented step-by-step in Table I, the BDT is trained and tested on separate event samples, and the sensitivity curves are obtained from parameter scans over first-principles cross sections rather than from fits to data. These features make the analysis internally consistent and the central claim concrete, provided the detector-performance assumptions, especially muon charge identification, are quantitatively justified.","major_comments":[{"comment":"The assumption of perfect muon sign identification is load-bearing and currently unquantified. The background suppression to 2.23 events after the full BDT chain rests on selecting a μ+ above 100 GeV and vetoing all μ− above 30 GeV. With 1.89×10^7 CC ν_μ events per year in the 10 kg detector, a charge-misidentification rate of 10^-3 would inject roughly 1.9×10^4 fake signal candidates before kinematic cuts, and a rate of 10^-4 would inject roughly 1.9×10^3, comparable to or larger than the 6.93×10^3 baseline background after the μ+ and μ− requirements. The manuscript states, 'For simplicity, we assume perfect muon sign identification within this setup and above the energy threshold,' and emphasizes that sign identification is crucial, but it does not assign a misidentification rate, does not show that the Evis, pT-Δφ, charm-tag, and BDT selections remove such fakes, and does not train or evaluate the BDT on a mis-ID-inclusive sample. Since the quoted λ bounds and the claimed sensitivity to thermal freeze-out and freeze-in targets depend on a final background of O(2) events, the authors should either provide a realistic charge-misidentification model for the proposed magnetized spectrometer or derive the maximum tolerable mis-ID rate as a function of the BDT cut and demonstrate that this requirement is compatible with the detector concept.","section":null}],"minor_comments":[{"comment":"The 'All Events' row appears to have missing entries: for seven numerical columns it lists only five numbers, so the NC displaced event rate (and possibly the 'All' total) is absent. Please complete the table so that the initial background totals are transparent and sum consistently with the later rows.","section":null},{"comment":"The 10 ton detector projection is described only by the statement that a stronger BDT cut of >0.90 is used and that the background after the BDT cut is O(100) events. A table analogous to Table I for the 10 ton detector, or a clear scaling argument from the 10 kg results, would make the advertised order-of-magnitude improvement in Fig. 1 verifiable.","section":null},{"comment":"The sensitivity criterion is given as S/√B > 2 with S > 3, but the curves are not labeled with a confidence level or a statistical procedure (e.g., Poisson exclusion with nuisance parameters). Please state the statistical interpretation explicitly.","section":null},{"comment":"There are several typos and style inconsistencies: 'neutrophilic' appears where 'neutrinophilic' is meant in Sec. I, and 'asssumed' appears instead of 'assumed' in Sec. III in the discussion of the hadronic calorimeter. A careful proofread would remove these.","section":null},{"comment":"The statement that a 1.5 TeV muon bends by approximately 2 cm in a 10 m, 1 T magnet is not by itself a demonstration of sign identification; the charge-sign resolution is set by the sagitta measurement and hit resolution. A sentence clarifying the assumed sagitta precision and its implication for the charge-misidentification rate would strengthen the detector-performance discussion.","section":null}],"recommendation":"major_revision","confidential_remarks":"The paper fits the journal's scope and the central physics case is attractive. The main issue is the unmodeled muon charge-misidentification background, which is explicitly acknowledged in the text as crucial but never quantified; because the quoted sensitivity and the connection to thermal DM targets depend on the final background of 2.23 events, this needs to be addressed before the projection can be regarded as robust. The rest of the analysis appears sound, and the stepwise Table I and the separate training/testing of the BDT are commendable. I do not see concerns about citation practice or novelty disclosure."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a well-executed sensitivity study, the first to apply the mono-neutrino wrong-sign muon process to a forward neutrino detector at a 3 TeV muon collider. The model and relic targets are from earlier papers, including by the same authors, but the application to MuColnu—flux, backgrounds, BDT, sensitivity curves for 10 kg and 10 ton—is new and done carefully. Table I is transparent: they show how each cut reduces CC/NC prompt and displaced backgrounds, and they validate the signal cross section against semi-analytic estimates. That is the kind of accounting a referee wants. The invisible Higgs recast is a straightforward bonus, not the main event.\n\nThe main weak spot is exactly the one the paper flags: muon sign identification is assumed perfect above threshold. The final background is 2.23 events after the BDT, but the baseline selection (E_mu+ > 100 GeV, E_mu- < 30 GeV) leaves 6.93e3 events. With ~1.9e7 CC nu_mu events per year, a true mu- reconstructed as mu+ at even 1e-3 would inject ~1.9e4 candidates, many with high reconstructed energy and no true mu+ to veto. The subsequent kinematic cuts and BDT would likely remove most of these, but the paper does not demonstrate that, and the BDT was not trained on a mis-ID-inclusive sample. This is a load-bearing detector-performance assumption, not an internal inconsistency. For a 10 kg detector the claim that even a 10 kg-yr exposure reaches thermal relic targets depends on it.\n\nEverything else is minor: the neutrino flux comes from a private communication (they thank Fieg and Wulzer), so the quantitative reach is not independently reproducible without the flux files; systematic uncertainties on the cross section are neglected but argued to be a few percent; charm tagging is an assumed 80% efficiency. These are the normal caveats of a forecast.\n\nI would send this to a serious referee. The paper is honest, the background treatment is the most detailed I have seen for this channel at a muon collider, and it gives the MuCol community a concrete target for detector design. The referee should ask for a mis-ID curve or a statement of the required charge ID efficiency, and ideally the flux files, but the physics case holds up.","headline":"A careful sensitivity forecast for a wrong-sign muon search at a forward MuCol neutrino detector; the physics case is solid, but the claimed reach depends on an unquantified muon charge mis-ID rate that needs a number before the bound can be trusted.","tokens_in":18397,"tokens_out":2513,"would_cite":true,"duration_ms":21930,"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":"A 10 kg neutrino detector downstream of a future muon collider could reach neutrino-portal dark matter.","keywords":["neutrino-portal dark matter","muon collider","neutrinophilic scalar mediator","mono-neutrino scattering","wrong-sign muon search","thermal freeze-out dark matter","freeze-in sterile neutrino dark matter","forward neutrino detector"],"falsifier":"Measure the charge-misidentification rate for high-energy $\\mu^-$ in the proposed magnetized spectrometer (10 m, 1 T) with a 1.5 TeV test beam: if the rate exceeds roughly $10^{-4}$, the ~$3\\times10^4$ CC events with an energetic $\\mu^-$ that the analysis rejects would produce several fake signal events per year, comparable to or larger than the ~2 surviving background events, and the claimed $\\lambda_{\\mu\\mu}$ limits would not hold. A full detector simulation that includes charge mis-ID and pion misidentification, run through the same BDT, would either reproduce the projected $S/\\sqrt{B}>2$ reach or show it degrade below the discovery threshold.","tokens_in":17294,"feed_emoji":"⚡️","tokens_out":12906,"duration_ms":105031,"temperature":0.7,"pith_summary":"The paper's central proposal is that a future 3 TeV muon collider is also a uniquely clean source of high-energy neutrinos: stored muons decay into a nearly pure, tightly collimated beam of $\\nu_\\mu$ and $\\bar{\\nu}_e$, with flux uncertainties far below those of proton-sourced neutrino beams. The authors argue that a compact detector (\"MuCol$\\nu$\") placed about 100 m downstream of the interaction point can exploit this beam to search for a neutrinophilic scalar $\\phi$ that couples to neutrinos and dark matter, using the mono-neutrino process $\\nu_\\mu N \\to \\mu^+ \\phi X$ with a wrong-sign positive muon in the final state. Their central claim is that even a 10 kg-year exposure, analyzed with kinematic cuts and a boosted decision tree, can probe values of the neutrino–$\\phi$ coupling that reach the relic-abundance targets for thermal freeze-out (Dirac fermion and complex scalar) dark matter and for freeze-in sterile-neutrino dark matter, for $\\phi$ masses from roughly 1 MeV to 10 GeV, surpassing the projected reach of DUNE, FLArE, and IceCube-Gen2. If this is right, neutrino-portal dark matter—one of the least constrained ways dark matter could couple to ordinary matter—could be tested with a detector roughly the size of a shoebox in the first year of a future muon collider.","feed_headline":"10 kg detector could probe neutrino-only dark matter","feed_subtitle":"Muon decays create an ultra-clean neutrino beam; one year of data would reach freeze-out and freeze-in dark matter targets.","key_machinery":"The central object is the lepton-number-carrying scalar $\\phi$—the neutrinophilic mediator—produced through $\\nu_\\mu N\\to\\mu^+\\phi X$; because $\\phi$ escapes the detector and carries lepton number, the event appears as an apparent lepton-number-violating process with a positively charged muon and missing energy. The argument is carried by the combination of a nearly pure $\\nu_\\mu$ beam (with the corresponding anti-$\\nu_\\mu$ directed away from the detector), a magnetized spectrometer that assigns the muon charge sign, a set of energy and transverse-momentum discriminants, and a boosted decision tree that reduces neutrino-induced backgrounds by roughly six orders of magnitude while keeping 20–40% of signal events. For small $\\phi$ masses the cross section scales approximately as $\\sigma\\simeq 10^{-37}\\,\\mathrm{cm}^2 \\times \\lambda^2 \\times (E_\\nu/\\mathrm{TeV})$, so the projected sensitivity is driven directly by the large, well-understood muon-collider neutrino flux rather than by a large detector.","core_discovery":"The discovery claim is that the $\\nu_\\mu N\\to\\mu^+\\phi X$ mono-neutrino process, recognized by an apparent lepton-number-violating wrong-sign $\\mu^+$ without an accompanying energetic $\\mu^-$, is observable above Standard Model backgrounds in a 10 kg detector sitting in the forward neutrino beam of a 3 TeV muon collider. With the selection described—$E_{\\mu^+}>100$ GeV, no $\\mu^-$ above 30 GeV, $E_{\\mu^+}/E_{\\text{vis}}>0.5$, a charm veto, transverse-momentum correlations, and a BDT cut—the paper projects that backgrounds fall by more than six orders of magnitude while signal efficiency remains at 20–40%, yielding sensitivity to $\\lambda_{\\mu\\mu}$ values as low as about $10^{-2}$ with the 10 kg detector and about $10^{-3}$ with a 10 ton detector in the few-hundred-MeV mass range. At these couplings, the model can simultaneously explain the thermal relic abundance of Dirac-fermion or complex-scalar dark matter, or populate sterile-neutrino dark matter through enhanced neutrino self-interactions. The paper further recasts Higgs-to-invisible decays at the muon collider to bound $\\lambda_{\\mu\\mu} \\gtrsim 0.09$, confirming that the neutrino-scattering channel is the more sensitive one.","pith_inferences":["Beyond the paper, the same wrong-sign-muon channel would also be a clean probe of neutrino trident production at TeV energies, which the paper mentions as a guaranteed physics case, and of nuclear parton distributions at high $x$.","Beyond the paper, the unmodeled charge-misidentification rate is the most direct way the projection could fail; a test of the magnetized spectrometer with a 1.5 TeV muon beam would determine whether the required $\\mu^+$–$\\mu^-$ separation is achievable.","Beyond the paper, a null 10 kg-year result would exclude the neutrinophilic-scalar explanation of the relic abundance over most of the 1 MeV–10 GeV window, pushing neutrino-portal models toward heavier or more weakly coupled regions that would need a higher-energy muon collider.","Beyond the paper, this analysis assumes only the $\\lambda_{\\mu\\mu}$ coupling is nonzero; a future detector with tau identification could use the same neutrino beam to search the $\\lambda_{\\tau\\tau}$ channel."],"forward_implications":["A 10 kg-year exposure is enough to reach the thermal freeze-out relic-abundance lines for Dirac-fermion and complex-scalar dark matter and the freeze-in sterile-neutrino target for $\\phi$ masses from about 1 MeV to 10 GeV.","For a 10 ton detector (equivalently 1 ton operating for 10 years), the projected coupling limits improve by roughly an order of magnitude, reaching $\\lambda_{\\mu\\mu} \\lesssim 10^{-3}$ in the few-hundred-MeV range.","In the mass range where DUNE, FLArE, and IceCube-Gen2 are projected to have their best sensitivity, MuCol$\\nu$ is projected to be more sensitive or comparable, and it exceeds the reach of invisible-Higgs searches at the HL-LHC and at the muon collider itself.","The search requires only a compact detector whose 10 kg target mass is similar to existing forward neutrino detectors at the LHC, so the necessary instrumentation is close to demonstrated designs.","Because the neutrino flux comes from well-understood muon decays, the flux systematic uncertainties are much smaller than for proton-sourced neutrino beams, so the projected limit is not dominated by flux-normalization error."],"supporting_citations":[{"why":"Supplies the 3 TeV muon-collider beam parameters and the neutrino flux that set all event rates.","marker":"[1]"},{"why":"Introduces lepton-number-charged scalars and the $\\nu_\\mu N\\to\\mu^+\\phi X$ mono-neutrino process used as the signal.","marker":"[4]"},{"why":"Defines the Dirac-fermion and complex-scalar thermal freeze-out dark-matter targets and the DUNE sensitivity baseline.","marker":"[5]"},{"why":"Provides the signal and background event-generation framework and the FLArE projected sensitivity used for comparison.","marker":"[18]"},{"why":"Gives the $h\\to\\nu\\nu\\phi$ decay width expressions used to recast invisible-Higgs constraints on the mediator.","marker":"[20]"},{"why":"Proposes the compact magnetized-spectrometer neutrino detector concept that MuCol$\\nu$ builds on.","marker":"[28]"},{"why":"Used to compute the signal cross sections with the neutrino-nucleon parton-level model.","marker":"[34]"},{"why":"Supplies the nCTEQ15 nuclear parton distribution functions used for the iron-target cross sections.","marker":"[35]"},{"why":"Models the neutrino-induced backgrounds, including charm production and in-flight pion and kaon decays.","marker":"[36]"},{"why":"Provides the projected muon-collider precision on invisible Higgs decays used for the complementary bound.","marker":"[47]"}],"fun_headline_variants":["10 kg detector at muon collider could reveal neutrino-portal dark matter","Neutrino-only dark matter: 10 kg detector at a muon collider","Muon collider's forward neutrinos may reveal dark matter in 10 kg detector","Tiny 10 kg detector could spy neutrino-portal dark matter at muon collider","Neutrino-portal dark matter probe: 10 kg detector at muon collider"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole projected reach depends on perfect muon sign identification for every muon above the energy thresholds, with no charge-misidentification rate modeled; even a tiny probability of misreading a high-energy negative muon as positive would put ordinary charged-current events into the signal sample and erase the six-order-of-magnitude background suppression.","fun_headline_variants_meta":{"raw":{"variants":["10 kg detector at muon collider could reveal neutrino-portal dark matter","Neutrino-only dark matter: 10 kg detector at a muon collider","Muon collider's forward neutrinos may reveal dark matter in 10 kg detector","Tiny 10 kg detector could spy neutrino-portal dark matter at muon collider","Neutrino-portal dark matter probe: 10 kg detector at muon collider"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000893,"raw_usage":{"total_tokens":3933,"prompt_tokens":1113,"completion_tokens":2820,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":729,"completion_tokens_details":{"reasoning_tokens":2709}},"tokens_in":729,"tokens_out":2820,"duration_ms":19523,"temperature":1.0,"reasoning_tokens":2709,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T15:58:07.025965+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the charge-misidentification rate for high-energy $\\mu^-$ in the proposed magnetized spectrometer (10 m, 1 T) with a 1.5 TeV test beam: if the rate exceeds roughly $10^{-4}$, the ~$3\\times10^4$ CC events with an energetic $\\mu^-$ that the analysis rejects would produce several fake signal events per year, comparable to or larger than the ~2 surviving background events, and the claimed $\\lambda_{\\mu\\mu}$ limits would not hold. A full detector simulation that includes charge mis-ID and pion misidentification, run through the same BDT, would either reproduce the projected $S/\\sqrt{B}>2$ reach or show it degrade below the discovery threshold.","supporting_citations":[],"review_version":1}