{"id":"934c077a-1e20-4c84-86cd-9409bb8a92ad","arxiv_id":"2510.27406","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"A nine-sector muon veto with single-sector trigger logic is projected to reject 99.7% of muon-induced events in the CROSS 0νββ region of interest, reducing that background to ~2e-3 cnts/keV/kg/yr.","lead":"The CROSS double-beta decay experiment installed a nine-sector plastic-scintillator muon veto around its cryostat in the Canfranc underground lab. The paper shows through simulations and a two-detector test that the veto can reduce muon-induced background near the signal energy to an acceptable ~2e-3 counts/keV/kg/yr, at the cost of ~18% dead time.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The headline 2e-3 cnts/keV/kg/yr rests on Table 1's 1 MeV veto threshold, but the adopted configuration uses 1.6–1.7 MeV thresholds, so the quoted background may not match the installed settings.","rationale":"The reader's weakest assumption is the Hall A muon flux/angular distribution being used for Hall B. That is a real but secondary issue: a modest flux difference would scale the residual background roughly linearly and, unless extreme, would not overturn the qualitative conclusion. A more direct and internal concern is that Table 1, the source of the abstract's 2e-3 cnts/keV/kg/yr, assumes a 1 MeV veto threshold, while the configured trigger rates in Sec. 4.3 correspond to 1.7/1.6 MeV thresholds. Because the muon detection efficiency is threshold-dependent and the paper does not quote a Table-1-equivalent background index for the actual thresholds, the central number may not correspond to the hardware settings. This is not a disagreement with the physics or an attack on the authors; it is a request for the missing calculation. The RUN13 validation does not directly test the ROI residual after all cuts, so it cannot close this gap. A dedicated MC rerun with the installed thresholds would settle the concern quickly. I therefore keep the reader's CONDITIONAL verdict: the engineering is plausible, but the headline background index needs to be recomputed and reported for the configuration actually deployed.","tokens_in":19547,"tokens_out":7621,"duration_ms":87511,"concrete_test":"Re-run the full CROSS Geant4 simulation with the installed threshold configuration of Sec. 4.3/5.1 — Lateral modules at 1.7 MeV, Bottom at 1.6 MeV, Top at 1 MeV — and reproduce Table 1 with the same multiplicity and light-detector cuts. Compare the MV>0 row to 1.8(1)e-3 cnts/keV/kg/yr, and also fold in the ~18% dead time. If the recomputed value exceeds ~2e-3, the abstract's headline background reduction is tied to a threshold setting that was not adopted.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central background claim is taken from Table 1, which is explicitly computed with a 1 MeV threshold in every veto module. The configuration actually installed (Sec. 4.3) sets module rates to 0.7 Hz (Lateral) and 0.3 Hz (Bottom), corresponding to energy thresholds of 1.7(1) MeV and 1.6(1) MeV; only the Top remains near 1 MeV. The veto's tagging efficiency is strongly threshold-dependent: Fig. 14 shows the muon-induced energy-deposit distribution in the veto rising from ~1.5 MeV, so raising the threshold from 1 MeV to 1.6–1.7 MeV removes part of the muon bump. The paper never reports the ROI background index for the actual 1.7/1.6 MeV thresholds. The quoted ~18% dead time corresponds to the actual settings, while the quoted ~2e-3 cnts/keV/kg/yr corresponds to the 1 MeV settings. If the residual BI at the installed thresholds is appreciably larger than 2e-3, the central claim is weaker than presented. The RUN13 validation (Table 4) with the actual thresholds finds only 85–88% coincidence tagging for E>3 MeV events; the 99.7% total rejection is an unvalidated combination of the multiplicity cut and veto efficiency.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper reports on the design, construction, operation, and optimization of a muon veto system for the CROSS 0νββ decay search at the Canfranc Underground Laboratory. The veto consists of four lateral, one top, and four bottom sectors based on polystyrene scintillator bars read out by SiPMs or PMTs. Geant4 simulations using the measured Hall A muon flux and angular distribution are used to predict muon-induced backgrounds in the CROSS detector. The simulations are validated against the rate of muon-like events in LMO bolometers (2.5(4) vs 2.6(1) cnt/d/LMO) and against a two-module cryogenic test (RUN13), where the coincidence tagging efficiency at E>3 MeV agrees after a 17% trigger-bit correction (85% vs 88%). The optimization of trigger rates is performed using a sensitivity metric that balances dead time and background index. The paper concludes that the adopted single-sector trigger logic rejects 99.7% of muon-induced events in the ROI, reducing the background to ~2×10^-3 cnts/keV/kg/yr, with a dead time of ~18%.","tokens_in":19860,"tokens_out":9633,"duration_ms":77972,"significance":"If the reported performance holds, the CROSS muon veto would reduce the dominant cosmic-muon background to a level compatible with a high-sensitivity 0νββ search. The strengths of the paper are the careful description of the veto hardware and DAQ, the direct comparisons of simulations to measured muon rates and to a dedicated two-module cryogenic run, and the transparent presentation of the optimization procedure. The MC framework is partially validated, and the paper is honest about the lack of a Hall B flux measurement. However, the central quantitative claim—the 99.7% rejection and the ~2×10^-3 cnts/keV/kg/yr background index—is not consistently tied to the actual installed threshold settings, which is a load-bearing issue that must be addressed before the result can be accepted as stated.","major_comments":[{"comment":"The abstract and conclusions quote a rejection of 99.7% and a residual background of ~2×10^-3 cnts/keV/kg/yr. These numbers come from Table 1, which explicitly assumes a 1 MeV energy threshold in every veto module. The configuration actually adopted after the optimization of §4.3 sets Lateral modules at 0.7 Hz (~1.7 MeV) and Bottom modules at 0.3 Hz (~1.6 MeV), with Top near 1 MeV. The paper never reports the background index or rejection efficiency for these installed thresholds. Since the muon-induced energy-deposit spectrum in the veto modules begins to rise at ~1.5 MeV (Fig. 14), raising the threshold to 1.6–1.7 MeV is expected to reduce the tagging efficiency and increase the residual background. The sensitivity optimization of §4.3 uses the correct thresholds via Eq. 4.1, but the BI at the optimum is not quoted. Please provide the BI and rejection efficiency for the actual installe","section":"§3.6.2, Table 1; §4.3"},{"comment":"The validation of the veto efficiency is carried out with a two-module CUPID-Mo setup at E>3 MeV, yielding 85% (corrected) experimental efficiency versus 88% simulated. This is a useful check, but the 99.7% rejection claimed for CROSS is a Monte Carlo projection for the full 42-crystal array that combines the veto with the multiplicity cut and the light-detector cut; this combination is not directly validated. In addition, during RUN13 six SiPMs were unstable and turned off (§5.1), and the effect of these dead channels on the coincidence efficiency is not quantified. Please estimate the impact of the dead channels on the achieved rejection and discuss any available cross-check of the full-array simulation (e.g., the 10-crystal RUN9 data) against the measured coincidence rates.","section":"§5.3, Table 4"}],"minor_comments":[{"comment":"Typo: 'while he other is identified' should read 'while the other is identified'.","section":"§4.2"},{"comment":"Typo: 'one order of magnitude higher that' should read 'higher than'.","section":"§3.6.2"},{"comment":"Duplicate definite article: 'An RMS of the the Light-to-Heat distribution'.","section":"Table 3"},{"comment":"The 17% correction factor for the disabled trigger bit is applied uniformly to coincidence rates. Please justify that the missed triggers are uncorrelated with the LMO events, or provide an uncertainty on this correction.","section":"§5.1"},{"comment":"The muon flux and angular distribution from Hall A are used for Hall B, which is ~20 m closer to the Rioseta valley. The paper notes the expected increase but does not quantify the effect on the background index. A short systematic study (e.g., scaling the flux by a few percent) would strengthen the conclusions.","section":"§3.2"}],"recommendation":"major_revision","confidential_remarks":"The paper is a solid instrumentation manuscript for JINST, with a detailed description of a functioning muon veto and a credible partial validation. The main issue is that the headline background index is computed for a 1 MeV threshold while the installed system operates at ~1.6–1.7 MeV; the paper must report the actual BI for the adopted settings. This is a fixable issue, but it is central to the paper's claim, so I recommend major revision rather than minor. I do not see any reason to doubt the integrity of the work or the validity of the simulation framework."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's my take. This is a competent, useful instrumentation paper for CROSS and the broader cryogenic rare-event community. The new content is the nine-sector veto geometry, the comparison of two-sector vs single-sector trigger logics, and the dead-time/background optimization. The calibration procedure with a 60Co source and the rate-vs-threshold fits are straightforward and adequate. The simulation is validated in two places: the LMO muon rate (2.5(4) vs 2.6(1) cnt/d/LMO) and the RUN13 coincidence efficiency after the trigger-bit correction (85% vs 88% at >3 MeV). That gives me reasonable confidence the MC is roughly right. The soft spots are real but not fatal. First, the headline 2e-3 cnts/keV/kg/yr and 99.7% rejection come from Table 1, which assumes a 1 MeV veto threshold. The installed configuration uses 0.7 Hz Lateral and 0.3 Hz Bottom rates, corresponding to 1.7(1) and 1.6(1) MeV thresholds. The paper never reports the background index at those thresholds. The optimization section says they simulated BI for each rate set, but they don't give the value for the chosen operating point. The abstract and conclusions repeat the 1 MeV number as if it were the design number. That matters because Fig. 14 shows the muon bump starts around 1.5 MeV, so the threshold choice sits right on the edge of the muon signal. If the true BI at 1.6–1.7 MeV is appreciably above 2e-3, the central claim weakens. I suspect the authors know the number—they just didn't write it down. Second, the 99.7% rejection is a combined MC result including the multiplicity cut and veto. The measured validation is only the coincidence tagging efficiency for two LMOs, at 85% corrected vs 88% simulated. That's consistent, and the multiplicity cut is a separate effect that has been measured in other CROSS runs, so I'm not alarmed. But a reader should not confuse the validated 85–88% with the projected 99.7%. Third, the Hall A muon flux and angular distribution applied to Hall B is an acknowledged assumption. Since the veto design and BI projection depend on it, that's a source of systematic uncertainty that is not quantified. Minor, because it's the best data available and the authors say so. Overall: solid instrumentation paper that deserves a serious referee. The main request should be to report the background index and rejection efficiency at the installed thresholds, and to add a systematic uncertainty on the MC projection. The engineering contribution is real and the validation, while partial, is honest. I'd send it to review.","headline":"Solid, useful instrumentation paper let down by a reporting gap: the headline background index is computed at a 1 MeV veto threshold, while the installed configuration runs at 1.6–1.7 MeV, and that interior number is never shown.","tokens_in":760,"tokens_out":1265,"would_cite":true,"duration_ms":52593,"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":"The CROSS experiment's new muon veto, run in single-sector trigger mode, rejects 99.7% of muon-induced events in the 100Mo neutrinoless double-beta decay region of interest, leaving about 2e-3 counts/keV/kg/yr of muon-induced background.","keywords":["muon veto","neutrinoless double-beta decay","100Mo","CROSS experiment","background rejection","Monte Carlo simulation","cryogenic bolometers","scintillation detectors"],"falsifier":"Measure the muon flux and angular distribution at the CROSS location using the same muon-monitor technique applied to the neighboring hall; if the integrated flux is more than about 10–20% higher than assumed, or the dominant arrival direction shifts, the stated 99.7% rejection and 2e-3 counts/keV/kg/yr background index would have to be re-derived. In the longer term, the first background run of the full CROSS array should observe a rate of muon-veto-coincident events in the region of interest consistent with that prediction.","tokens_in":19384,"feed_emoji":"🛡️","tokens_out":4760,"duration_ms":44762,"temperature":0.7,"pith_summary":"This paper reports the design, construction, and validation of a dedicated muon veto for the CROSS cryogenic experiment, which searches for neutrinoless double-beta decay of 100Mo at the Canfranc underground laboratory. The authors aim to show that the residual cosmic-ray muon flux, the dominant background near the 3 MeV region of interest, can be suppressed to a level compatible with the experiment's target sensitivity. By combining Monte Carlo simulations with measurements of calibrated veto modules and a two-detector cryogenic test run, they establish that triggering on a single veto sector and rejecting all coincident thermal-detector events reduces the muon-induced background to about 2e-3 counts/keV/kg/yr, rejecting 99.7% of muon-induced events. The work matters because it turns a known dominant background into a manageable one, preserving CROSS's projected sensitivity.","feed_headline":"Muon veto rejects 99.7% of muon events in CROSS double-beta search","feed_subtitle":"Anticoincidence with any of nine detector sectors cuts residual muon background to 2e-3 counts per keV per kg per year.","key_machinery":"The load-bearing mechanism is the time-correlated veto: the muon veto consists of nine sectors (four lateral, one top, and four bottom) made of polystyrene scintillator bars read out by silicon photomultipliers or photomultipliers, and each thermal-detector event is rejected if any sector registers an energy deposit above a calibrated threshold within a ±1–2 ms coincidence window. The performance is quantified with Monte Carlo simulations of muons crossing the setup, calibrated module by module with a 60Co source, and validated against low-temperature bolometer measurements in the actual cryogenic facility.","core_discovery":"The central claim is that the adopted muon-veto strategy—an OR of all nine plastic-scintillator sectors, with a 2 ms coincidence window against the thermal detectors and a single-hit multiplicity cut—reduces muon-induced background in the 2.7–3.4 MeV region of interest to about 2e-3 counts/keV/kg/yr, an acceptable level for a high-sensitivity neutrinoless double-beta decay search. This is demonstrated through dedicated Monte Carlo simulations using the measured muon angular distribution and rock-overburden parameterization, validated against stable veto trigger rates and against a 555 h run with two CUPID-Mo bolometers, where measured coincidence efficiencies agree with simulation after corr","pith_inferences":["If a direct muon flux measurement at the CROSS location in Hall B shows a flux higher than the Hall A parameterization used here—the paper notes Hall B is about 20 m closer to the Rioseta valley—the projected 2e-3 counts/keV/kg/yr background could rise proportionally, so a Hall B measurement would be the most direct stress test of the margin.","The 18% dead time from vetoing on any of nine sectors suggests a testable refinement: module-level rather than sector-level veto patterns, or energy-dependent veto thresholds, could reduce dead time while preserving rejection efficiency.","Because the coincidence window is tied to the slow thermal signal, using the faster light-detector trigger position (as done in this analysis) may allow shorter windows and lower dead time in future operation.","The calibration and optimization protocol—per-module energy calibration, background-rate fitting, and sensitivity maximization—can be transferred directly to other underground cryogenic rare-event searches facing similar muon fluxes."],"forward_implications":["Muon-induced background in the CROSS region of interest is reduced to about 2e-3 counts/keV/kg/yr, compatible with the experiment's high-sensitivity goal.","The adopted single-sector trigger logic suppresses 99.7% of muon-induced events, compared with 99.2% for a two-sector coincidence trigger and 98.7% for the thermal-detector multiplicity cut alone.","The resulting dead time of about 18% is an accepted trade-off that preserves the experiment's live-time-weighted sensitivity.","Validation with two CUPID-Mo modules shows measured muon-veto rejection efficiencies of 85–90% for events above 3–10 MeV, matching simulation and supporting the projected performance for the full CROSS array.","The veto system operates stably with an average trigger rate varying by less than 2% over about 100 h, confirming the slow-control threshold-stabilization approach."],"fun_headline_variants":["CROSS muon veto: 99.7% rejection of muon background","Muon veto lowers CROSS background to acceptable 2e-3","99.7% muon cut: CROSS veto validated in 555-hour run","CROSS veto removes 99.7% muon events for double-beta"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The load-bearing assumption is that the muon flux and angular distribution measured in one hall of the Canfranc underground laboratory, where the rock overburden is slightly thicker, also describe the adjacent hall where CROSS is installed; no direct muon measurement exists at the CROSS location.","fun_headline_variants_meta":{"raw":{"variants":["CROSS muon veto: 99.7% rejection of muon background","Muon veto lowers CROSS background to acceptable 2e-3","99.7% muon cut: CROSS veto validated in 555-hour run","CROSS veto removes 99.7% muon events for double-beta"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.0009,"raw_usage":{"total_tokens":3794,"prompt_tokens":912,"completion_tokens":2882,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":656,"completion_tokens_details":{"reasoning_tokens":2797}},"tokens_in":656,"tokens_out":2882,"duration_ms":19865,"temperature":1.0,"reasoning_tokens":2797,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-04T06:57:09.681077+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the muon flux and angular distribution at the CROSS location using the same muon-monitor technique applied to the neighboring hall; if the integrated flux is more than about 10–20% higher than assumed, or the dominant arrival direction shifts, the stated 99.7% rejection and 2e-3 counts/keV/kg/yr background index would have to be re-derived. In the longer term, the first background run of the full CROSS array should observe a rate of muon-veto-coincident events in the region of interest consistent with that prediction.","supporting_citations":[],"review_version":1}