{"id":"e768539f-d5c2-4433-9ec1-fd2edf5e9cc7","arxiv_id":"1908.08299","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"A new IceCube event selection lowers its neutrino threshold below 1 GeV, and the first search for GeV neutrinos from compact binary mergers finds no excess and sets upper limits.","lead":"The IceCube Collaboration found no GeV neutrinos within seconds or minutes of ten compact binary merger detections, and set first limits in the 0.5 to 5 GeV range. The paper also validates an event selection that lowers IceCube's effective threshold below 1 GeV, opening a new energy window for multi-messenger searches.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Quantitative GeV limits rest on simulated effective area with no data/MC validation; the printed fluence exponent also appears to be off by many orders of magnitude.","rationale":"Read in good faith, this is a proceedings-style proof-of-concept: the null results themselves are probably robust, and the idea of using DeepCore-style low-energy event selection for gravitational-wave multimessenger searches is plausible. However, the paper's novel contribution is quantitative sub-GeV sensitivity and first fluence limits, and that contribution is supported only by unvalidated simulation. The reader identified the same weakest assumption: the simulated detector response used to build the event selection and effective area is not checked against data. I agree with that assessment. The apparent exponent error in the headline fluence limit is a concrete symptom of the same problem: the numerical result has not been verified end-to-end. No formal verification or released code is provided, so independent reproduction is not currently possible. The appropriate outcome is conditional acceptance: the scientific idea is viable and the null observation is credible, but the quantitative limits and the '<1 GeV' claim should not be cited as final until a data/MC closure study or at least a systematic uncertainty band on the effective area is supplied. My stress-test therefore does not move the reader's verdict.","tokens_in":5201,"tokens_out":7248,"duration_ms":79855,"concrete_test":"Take roughly 100 days of IceCube data, excluding the [t-500s, t+500s] windows around LVC triggers, and process it through the final GeV selection. Compare the observed distributions of total charge, depth, and number of causally connected DOMs with the GENIE plus noise Monte Carlo prediction. Fit the noise normalization to the off-source sideband to obtain a data/MC closure factor; if that factor differs from unity by more than the statistical uncertainty, recompute the Section 3.1 upper limit with the closure-corrected effective area and restate the '<1 GeV' sensitivity claim. Separately, re-derive the fluence limit from the zero-event Poisson upper limit and the Figure 3 effective area to verify whether the printed exponent should be 10^-7 rather than 10^7.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative claim is the 0.5-5 GeV effective area and the derived fluence limits. Section 2.1 builds the final GeV event selection from GENIE 2.8.6 neutrino simulations plus a detector-noise simulation, reporting a >40% passing fraction below 5 GeV and a 0.02 Hz final rate, but no data/MC closure or systematic uncertainty is shown. Section 3.1 converts zero observed coincident events into an upper limit on fluence (1.84 x 10^7 neutrinos MeV^-1 cm^-2) by dividing by this simulated exposure. At 0.5-5 GeV, a neutrino is only a handful of causally connected DOM hits, so the selection is highly sensitive to modeled thermal/radioactive/scintillation noise and to GENIE cross sections and ice optics. If real noise correlations or the interaction model differ from simulation by even a factor of two, the effective area in Figure 3, and hence every quoted limit, changes by a comparable factor, because the analysis is a null count. Separately, the quoted 1.84 x 10^7 neutrinos MeV^-1 cm^-2 is unphysical for a detector with an effective area of order 10^3-10^4 cm^2 at GeV energies; a missing minus sign is likely, but as printed the central number cannot be reproduced from the text. Both issues indicate the quantitative result lacks a self-consistent, validated derivation.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This ICRC2019 proceedings paper presents a search for GeV-scale astrophysical neutrinos from compact binary mergers detected by LIGO/Virgo, using a new low-energy event selection in IceCube/DeepCore. The selection is built from GENIE 2.8.6 neutrino simulations and a detector-noise simulation, yielding the reported effective areas and passing fractions for electron and muon neutrinos. For the three BNS/NSBH mergers in the sample, zero events are found in the [t, t+3 s] prompt window, and the paper quotes an upper limit on the neutrino fluence in the 0.5-5 GeV range. An extended ±500 s search over BBH mergers yields another limit. The results are compared with Super-Kamiokande and high-energy neutrino constraints.","tokens_in":5460,"tokens_out":6883,"duration_ms":70392,"significance":"If the quantitative results are correct, this is an interesting proof of principle that IceCube can be sensitive to sub-GeV/GeV neutrinos, opening a new multi-messenger window. The paper has clear strengths: the null observation is robust, the assumed E^-2 source spectrum is stated, public LIGO/Virgo merger times are used, and the comparison with existing constraints is useful. However, the numerical limits are not reproducible from the text as printed, and the effective area and event-selection efficiency rest entirely on simulation with no data/MC closure or systematic uncertainties. The significance of the headline numbers is therefore not yet established, even though the new direction itself is valuable.","major_comments":[{"comment":"The GeV event selection and its efficiency rest entirely on GENIE 2.8.6 neutrino simulations and a detector-noise simulation; no data/MC closure or systematic uncertainty is shown. The final sample is noise-dominated before the final cuts (6 Hz of noise survives after the first stage, 0.2 Hz after applying causality, and 0.02 Hz final), so even a modest mismodeling of the noise correlations or of the low-energy light yield would scale the effective area in Figure 3. Because the search is a null count, every quoted fluence limit scales inversely with that effective area. Please add a data/MC comparison of the final event rate and of the distributions of the discriminating variables, or state explicitly that the quoted limits are preliminary and do not include this uncertainty.","section":"Section 2.1"},{"comment":"The headline limit, '1.84 x 10^7 neutrinos MeV^-1 cm^-2', cannot be reproduced from the information in the text, and the units are nonstandard. For a zero-count search over a 3 s window with an effective area of the magnitude implied by Figure 3, a 90% confidence upper limit on a differential fluence in the 500 MeV-5 GeV band should be orders of magnitude smaller; the printed value appears to have an exponent/sign error. Please provide the explicit limit formula (including the Poisson upper quantile, the assumed background rate, and the energy integral over the assumed E^-2 spectrum) and the numerical result of that calculation.","section":"Section 3.1"},{"comment":"The extended-window result is presented as an upper limit, but the procedure is not defined. The text states that the data recorded during GW170608 lie in the 5% tail of the background distribution and then quotes a limit of '50 x 10^3 neutrinos MeV^-1 cm^-2'; however, a single event in the 5% tail is not by itself an upper limit, and the same units/exponent problem as in Section 3.1 reappears. Please specify the counting or likelihood method, the confidence level, whether the observed event is included in the limit, and the numerical exposure used.","section":"Section 3.2"}],"minor_comments":[{"comment":"The claim to present the 'first search for GeV astrophysical neutrinos emitted from Compact Binary Mergers' should be qualified, because the Super-Kamiokande limit on GW170817 cited as [4] already covers part of the GeV range; suggest 'first search in IceCube' or 'first search below 1 GeV'.","section":"Abstract / Section 1"},{"comment":"The units 'neutrinos MeV^-1 cm^-2' are ambiguous for a fluence limit; please state whether this is a differential fluence per energy bin and, if so, specify the binning and whether the quoted number is a 90% confidence limit.","section":"Sections 3.1 and 3.2"},{"comment":"The caption says the present work is shown as a 'blue line' and the high-energy neutrino constraints as a 'blue shaded area'; using two blue entries in the same figure is confusing and should be fixed with distinct colors and/or labels.","section":"Figure 4"},{"comment":"The word 'subthreshold' is misspelled as 'subtreshold', and the table should indicate whether the Fermi-LAT candidate for GW170608 was confirmed as a counterpart or remains a sub-threshold candidate.","section":"Table 1"},{"comment":"Reference [14] should be 'Glück' rather than 'Glck', and the text should use the standard notation for the GRV98 parton distribution functions.","section":"References"},{"comment":"The effective area is shown for ν_e and ν_μ, but the prompt limit is said to be 'integrated over the three neutrino flavors'; please state the flavor assumption explicitly and note whether an effective area for ν_τ is included or neglected.","section":"Figure 3 / Section 3.1"}],"recommendation":"major_revision","confidential_remarks":"The paper is an ICRC proceedings contribution, so some brevity is expected, but the two quantitative results are currently not reproducible: one appears to contain an exponent error, and the other lacks a defined limit-setting procedure. The simulation-only validation of the effective area is also a substantive concern. These are correctable in a revision, and the direction is worth publishing, so I recommend major revision rather than rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This is a useful ICRC proceedings paper: it reports the first IceCube search for GeV astrophysical neutrinos from compact binary mergers, using a new event selection that pushes the threshold below 1 GeV. That is genuinely new. The selection logic is described clearly—cut on number of DOMs, causality, then topology—and the null result in the 3-second prompt window is honest and easy to read. The comparison with Super-K and high-energy limits is a nice way to show complementarity. Credit where due: the idea of lowering IceCube's energy reach this way and applying it to LIGO/Virgo events is a step forward, and the paper is upfront about it being a first pass.\n\nNow the soft spots. The quantitative limits are built on a simulated effective area from GENIE plus a detector noise simulation, with no data/MC closure, no systematic uncertainties, and no validation against real detector behavior. For a null count that is fine in principle, but it means the quoted limits are only as good as the simulation, and a factor of two in noise rate or cross section changes every number. That is a real limitation, though not disqualifying for a proceedings paper.\n\nMore concerning is the internal consistency of the printed limits. The prompt fluence limit is given as 1.84 x 10^7 neutrinos MeV^-1 cm^-2. That has to be wrong—a quick estimate using the effective area in Figure 3 gives something like 10^-7, so a missing minus sign is almost certain. The extended-window limit of 50 x 10^3 looks similarly suspicious. These are probably typos, but they are load-bearing numbers, and a reader cannot reproduce them from the text as written. That is a real check on the paper's reliability.\n\nWho is this for? People in multimessenger astronomy and low-energy neutrino detection. It is a signal that IceCube may be able to probe a new energy range, and it gives first limits that a future full analysis can improve on. It deserves a serious referee: the idea is sound, the result is new, and the flaws are fixable. I would send it to peer review, but with the expectation that the collaboration provides a real validation of the event selection on data and fixes the unit errors.\n\nBottom line: worth reading, worth citing, but treat the absolute limits with caution until the full analysis appears.","headline":"First IceCube search for GeV neutrinos from compact binary mergers, with a plausible new selection but limits that rest entirely on simulation and at least one printed number that looks like a typo.","tokens_in":5996,"tokens_out":2673,"would_cite":true,"duration_ms":29159,"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 first GeV neutrino search for compact binary mergers finds no signal and sets low-energy upper limits.","keywords":["neutrino astronomy","IceCube","GeV neutrinos","compact binary mergers","gravitational waves","multi-messenger astronomy","upper limits","DeepCore"],"falsifier":"Run the GeV event selection on a year of IceCube data and count events in the 1–5 GeV atmospheric neutrino band; the observed rate should match the roughly mHz prediction made from the simulation. A discrepancy beyond statistical uncertainty would show the effective area in Fig. 3 is not correct and would invalidate the quoted upper limits.","tokens_in":5022,"feed_emoji":"🔭","tokens_out":8443,"duration_ms":77216,"temperature":0.7,"pith_summary":"This proceedings paper describes the first search for astrophysical neutrinos in the 0.5–5 GeV energy range from compact binary mergers detected by LIGO and Virgo. The authors introduce a new event-selection method that effectively lowers IceCube's neutrino detection threshold from roughly 10 GeV to below 1 GeV by selecting compact, low-light events in the DeepCore subdetector and rejecting detector noise with causality and topology cuts. Applying this selection to seven mergers, they find no events in the three seconds following binary neutron star and neutron star–black hole mergers, yielding an upper limit of $1.84 \\times 10^{7}$ neutrinos MeV$^{-1}$ cm$^{-2}$ on the emitted fluence; a wider $\\pm 500$ s search around black hole mergers also shows no significant excess. If the method holds up, it opens a previously unobservable energy window for multi-messenger neutrino astronomy and complements existing TeV–PeV searches.","feed_headline":"No GeV neutrinos from compact binary mergers","feed_subtitle":"New IceCube selection drops below 1 GeV to set first upper limits on low-energy neutrino emission.","key_machinery":"The central mechanism is a low-energy neutrino event selection for IceCube/DeepCore, defined by a sequence of cuts: standard high-energy filters (1400 Hz to 15 Hz), an upper bound on the number of triggered optical modules, a causality requirement on hit pairs consistent with the speed of light in ice with scattering, and topology variables (interaction depth, local charge, total charge). This selection is what carries the argument; its simulated efficiency (Fig. 2) and effective area (Fig. 3) convert a null event count into a fluence upper limit via the assumed $E^{-2}$ spectrum.","core_discovery":"The central claim is that IceCube can be made sensitive to GeV neutrinos by a new selection of events in DeepCore. The selection first uses standard IceCube filters to reduce the atmospheric muon background from 1400 Hz to 15 Hz while retaining 98% of GeV neutrino events, then imposes a small number of causally connected optical modules and a causality condition between pairs of hits, cutting noise from 6 Hz to 0.2 Hz, and finally applies depth, charge, and event-topology cuts to reach a rate of 0.02 Hz. The resulting effective area is reported in Fig. 3, and the passing fraction exceeds 40% for neutrinos below 5 GeV, optimized between 1 and 2.5 GeV. The paper then applies this selection to the LVC catalog: for BNS and NSBH mergers, zero events are found in the [t, t+3s] window, which is interpreted as an upper limit of $1.84 \\times 10^{7}$ MeV$^{-1}$ cm$^{-2}$ integrated over 3 s in the 500 MeV–5 GeV range; for BBH mergers in a $\\pm 500$ s window, all events are consistent with background, with GW170608 in the 5% tail, leading to an upper limit of $5.0 \\times 10^{4}$ MeV$^{-1}$ cm$^{-2}$. The paper presents this as the first validation of a GeV neutrino search in IceCube.","pith_inferences":["Beyond the paper: because the event selection's efficiency comes entirely from simulation, a calibration sample of atmospheric neutrinos in the 1–5 GeV band would provide a direct check of the reported effective area; the paper does not present such a data/MC comparison.","Beyond the paper: the 3 s prompt window is tied to the 1.7 s delay of the GRB in GW170817; a source with delayed accretion or a longer-lived cocoon could emit GeV neutrinos over minutes, so the null result does not constrain those scenarios.","Beyond the paper: if the assumed $E^{-2}$ spectrum is wrong, the quoted fluence limits do not directly apply; for a softer spectrum, the sensitivity at the lower edge of the 500 MeV–5 GeV band would degrade, so the limit is spectrum-dependent.","Beyond the paper: applying this selection to a known gamma-ray burst with observed GeV photons would offer a test of the neutrino production mechanism: a correlation of neutrino count with GeV gamma fluence would support hadronic models."],"forward_implications":["IceCube's energy reach for astrophysical neutrino searches now extends below 1 GeV, making coincident low-energy neutrino searches possible for gravitational-wave and gamma-ray transient events.","The null prompt search places the first upper limits on GeV neutrino fluence from BNS/NSBH mergers, constraining hadronic acceleration and the matter environment around the merger site.","The upper limits complement, rather than compete with, Super-Kamiokande and high-energy neutrino telescope limits by probing a different energy band and thus different production mechanisms.","The same selection can be applied to future gravitational-wave events; as LIGO/Virgo sensitivity grows, the accumulated exposure will tighten these limits.","With KM3NeT and IceCube-Upgrade, the paper expects lower triggers and better reconstruction, straightforwardly improving GeV-regime sensitivity."],"supporting_citations":[{"why":"Provides the GENIE neutrino-interaction simulation used to compute the passing fraction and effective area of the GeV selection.","marker":"[10]"},{"why":"Supplies the detector noise simulation (thermal, radioactive, scintillation) used to estimate and suppress the dominant background in the GeV sample.","marker":"[15]"},{"why":"Defines the IceCube detector and its standard filters, which the GeV selection inherits and builds upon.","marker":"[5]"},{"why":"Documents the DeepCore subarray and its softer trigger that make sub-10 GeV neutrino detection possible.","marker":"[6]"},{"why":"Sets the previous GeV-regime neutrino limits from Super-Kamiokande on GW170817, the comparison benchmark for the new low-energy limits.","marker":"[4]"},{"why":"Provides the joint high-energy neutrino upper limits from ANTARES, IceCube, and Pierre Auger that the new GeV limits are compared against.","marker":"[18]"},{"why":"Supplies the conservative $\\pm 500$ s time window used for the extended search around binary black hole mergers.","marker":"[19]"},{"why":"Defines the initial LIGO/Virgo catalog of binary mergers used as the target list for the GeV neutrino search.","marker":"[1]"},{"why":"Reports the GW170817 detection and its 1.7 s gamma-ray delay, which motivates the 3 s prompt search window.","marker":"[17]"}],"fun_headline_variants":["IceCube's new low-energy trigger sees no neutrinos from mergers","First sub-GeV neutrino limits from binary mergers: all null","Zero GeV neutrinos from LIGO/Virgo mergers in new IceCube search","IceCube lowers threshold to GeV, finds no neutrinos from mergers"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The selection's efficiency and effective area come entirely from computer simulations of neutrino interactions and detector noise, and the paper shows no comparison of the selected GeV sample against real data; if those simulations misrepresent how IceCube responds at GeV energies, every fluence limit derived from the null observations would be wrong.","fun_headline_variants_meta":{"raw":{"variants":["IceCube's new low-energy trigger sees no neutrinos from mergers","First sub-GeV neutrino limits from binary mergers: all null","Zero GeV neutrinos from LIGO/Virgo mergers in new IceCube search","IceCube lowers threshold to GeV, finds no neutrinos from mergers"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000787,"raw_usage":{"total_tokens":3484,"prompt_tokens":968,"completion_tokens":2516,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":584,"completion_tokens_details":{"reasoning_tokens":2447}},"tokens_in":584,"tokens_out":2516,"duration_ms":19187,"temperature":1.0,"reasoning_tokens":2447,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:43:13.513718+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the GeV event selection on a year of IceCube data and count events in the 1–5 GeV atmospheric neutrino band; the observed rate should match the roughly mHz prediction made from the simulation. A discrepancy beyond statistical uncertainty would show the effective area in Fig. 3 is not correct and would invalidate the quoted upper limits.","supporting_citations":[{"cited_title":"Andreopoulos et al., Nucl","cited_arxiv_id":null,"evidence_quote":"Provides the GENIE neutrino-interaction simulation used to compute the passing fraction and effective area of the GeV selection."},{"cited_title":"Larson, Ph.D thesis, University of Alabama, Tuscaloosa (2013)","cited_arxiv_id":null,"evidence_quote":"Supplies the detector noise simulation (thermal, radioactive, scintillation) used to estimate and suppress the dominant background in the GeV sample."},{"cited_title":"Abbasi et al., Astropart","cited_arxiv_id":null,"evidence_quote":"Documents the DeepCore subarray and its softer trigger that make sub-10 GeV neutrino detection possible."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Sets the previous GeV-regime neutrino limits from Super-Kamiokande on GW170817, the comparison benchmark for the new low-energy limits."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the joint high-energy neutrino upper limits from ANTARES, IceCube, and Pierre Auger that the new GeV limits are compared against."},{"cited_title":"Baret et al., Phys","cited_arxiv_id":null,"evidence_quote":"Supplies the conservative $\\pm 500$ s time window used for the extended search around binary black hole mergers."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the initial LIGO/Virgo catalog of binary mergers used as the target list for the GeV neutrino search."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Reports the GW170817 detection and its 1.7 s gamma-ray delay, which motivates the 3 s prompt search window."}],"review_version":1}