{"id":"ef03f19d-f0cc-4c0b-b101-d3cd65e23b54","arxiv_id":"2507.08753","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The IceCube collaboration presents the combined multi-flavor ICEMAN sample and projects an improved galactic plane sensitivity with an expected 5.5 sigma Fermi pion-decay template significance.","lead":"This IceCube proceedings paper combines three neutrino event selections into one all-flavor sample, ICEMAN, and projects its sensitivity to the galactic plane. It reports over 20% improved discovery potential compared to the earlier single-sample analysis, with an expected 5.5 sigma signal under the Fermi pion-decay template.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Tables 1 and 2 appear mutually inconsistent: the quoted Fermi-pi0 discovery potential (18.7) is above the previously measured flux (~4.8) that the text says should yield 5.5 sigma, so the central sensitivity claim is not internally verifiable.","rationale":"The paper's purpose is to present ICEMAN as a more sensitive instrument, with the headline that injecting the previous best-fit galactic-plane flux yields a median local significance of 5.5 sigma (Table 2) and that sensitivities improve by more than 20%. I looked for the place where this claim is least secure. The most load-bearing step is the relationship between the quoted discovery potential and the expected significance. The text in Section 3 says the measured flux from [3] is above the flux needed for 5 sigma, and Table 2 then gives 5.5 sigma. But Table 1 lists the Fermi-pi0 discovery potential as 18.7 x 10^-12 TeV cm^-2 s^-1, while the [3] best-fit flux is approximately 4.8 x 10^-12 in those units. Under the standard definition (50% of pseudo-experiments to exceed 5 sigma), an injection at 4.8 would produce a median significance well below 5.5 sigma, not above it. Either the 18.7 entry is not the local 5 sigma threshold (e.g., it is trial-corrected/global), or the text's comparison to [3] is wrong, or the units in Table 1 differ from the 'best-fit flux' being injected. None of these resolutions are provided. This matters because the 5.5 sigma projection is the paper's central advertised result; if the definitions are inconsistent, the claim cannot be verified or reproduced from the text. This is independent of the reader's systematics concern: even with perfect cross-sample systematics, the claimed significance would be unsubstantiated until Tables 1 and 2 are reconciled. I therefore recommend keeping the existing CONDITIONAL verdict, but sharpening the conditions to require the authors to define local versus global significance and to demonstrate consistency between Table 1 and Table 2. I do not see this as grounds for rejection: the projected sensitivity method is standard, the event counts and livetimes are clearly reported, and the inconsistency may be a labeling or definition issue in a short proceedings paper. But it is the single most load-bearing concern I can identify.","tokens_in":9772,"tokens_out":24060,"duration_ms":251195,"concrete_test":"Re-run the Section 3 pseudo-experiments with the injected flux fixed to the value quoted in [3] in the same units as Table 1 (approximately 4.8 x 10^-12 TeV cm^-2 s^-1 per flavor at 100 TeV) and verify whether the median local significance is 5.5 sigma as in Table 2. Independently verify that injecting 18.7 x 10^-12 yields 50% of pseudo-experiments above the 5 sigma TS threshold as in Table 1. If both cannot hold simultaneously, Tables 1 and 2 are inconsistent; if they can, state explicitly whether the Table 1 threshold is local or trial-corrected and reconcile the text's comparison to [3].","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative claim — an expected median local significance of 5.5 sigma when injecting the previous best-fit Fermi-pi0 flux (Section 3, Table 2) — is not internally verifiable from the paper's own numbers. Table 1 quotes the Fermi-pi0 discovery potential as 18.7 x 10^-12 TeV cm^-2 s^-1, defined in the text as the flux for 50% of pseudo-experiments to exceed the TS corresponding to 5 sigma. The text then states that the flux measured in [3] is above the flux needed for 5 sigma. But the [3] best-fit flux, in the same units as Table 1, is about 4.8 x 10^-12 TeV cm^-2 s^-1, well below 18.7. If instead the injected 'best-fit flux' is a dimensionless multiple of the template (e.g., 4.8 times the nominal model), then Table 1's flux-at-100-TeV units are not the quantity being compared, and the text's comparison is undefined. The paper never states whether Table 1's 5 sigma threshold is local or trial-corrected, nor whether Table 2's 'local significance' corresponds to the same threshold. As printed, Tables 1 and 2 cannot both be correct under a single stated definition, so the headline 5.5 sigma projection and the 'more than 20% improvement' rest on an unstated and potentially inconsistent definition.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript introduces ICEMAN, a combined IceCube all-flavor dataset built from the DNN cascade, Enhanced Starting Track, and Northern Track event selections, and presents a template-based unbinned likelihood analysis of the Galactic plane using four gamma-ray-derived emission templates. It reports sensitivity and discovery-potential projections (Table 1) and an expected median local significance of 5.5σ when injecting the previous best-fit Fermi-π0 flux (Table 2). The abstract also advertises two PeVatron searches, one using a Cygnus Cocoon template and one correlating neutrinos with >100 TeV gamma-ray sources, but the body of the paper contains no results or method details for these searches; the conclusion defers them to future work.","tokens_in":10010,"tokens_out":10923,"duration_ms":122197,"significance":"If the sensitivity projection is correct, ICEMAN would provide a substantially more sensitive dataset for Galactic plane neutrino astronomy, and the claimed improvement of more than 20% over the previous cascade-only analysis is a concrete and useful benchmark. The likelihood construction (Eqs. 1-3) and the pseudo-experiment procedure are standard and consistent with prior IceCube analyses, and the paper is commendably clear about the dataset composition and overlap removal. However, the headline numbers are projections against an injected benchmark rather than a measurement, and the paper's own Tables 1 and 2 are not mutually consistent under a single stated definition of the 5σ discovery threshold. This internal inconsistency affects the central quantitative claim and needs to be resolved before the results can be assessed.","major_comments":[{"comment":"The central sensitivity claim is not internally consistent as printed. Table 1 lists a Fermi-π0 discovery potential of 18.7 × 10^-12 TeV cm^-2 s^-1, which the text defines as the flux needed for 50% of pseudo-experiments to exceed the TS corresponding to 5σ. The previous best-fit flux from [3], in the same units, is approximately 4.8 × 10^-12 TeV cm^-2 s^-1, which is below 18.7. The text nevertheless states that the [3] flux is above the 5σ threshold and that a 5σ pre-trial significance is likely, and Table 2 reports a median local significance of 5.5σ for that injected flux. The authors must state explicitly whether the 'discovery potential' in Table 1 is a local or trial-corrected 5σ threshold, and must explain how the values in Tables 1 and 2 can be reconciled under a single definition. Showing the TS distributions from the pseudo-experiments that produce the 5.5σ median would resolve this directly.","section":"Section 3, Tables 1 and 2"},{"comment":"No systematic uncertainties are included in the pseudo-experiment projection. The three sub-samples have different effective areas, angular resolutions, and energy resolutions, and the paper notes improved ice modeling and calibration only for the cascade reconstruction. If relative normalization, energy-scale, or ice-model uncertainties between the samples are neglected, the projected 5.5σ significance and the claimed >20% improvement may be optimistic. The manuscript should state explicitly whether systematics are neglected and should quantify the impact of representative systematic shifts, for example ±10% effective-area normalization or ±5% energy-scale shifts, on the median expected significance.","section":"Section 3, Eqs. (1)-(3) and Table 2"},{"comment":"The abstract states that the paper 'adopts two different analysis methods to search for galactic PeVatrons,' specifically a template-based approach to the Cygnus Cocoon region and a point-source hypothesis correlating IceCube neutrinos with >100 TeV gamma-ray sources. Sections 2-4 do not present either analysis: the method section describes only the Galactic plane template analysis, and the conclusion says such searches will be done in future work. The manuscript should either include these PeVatron analyses or revise the abstract and title so that they accurately reflect the content.","section":"Abstract and Conclusion"},{"comment":"The claim that 'the sensitivities have improved by over 20%' is not supported by a quantitative comparison in the manuscript. No prior sensitivity or discovery-potential values from [3] are quoted in the same units as Table 1, and it is unclear whether the comparison refers to the sensitivity row, the discovery potential row, or the expected median significance. Please provide the numerical comparison for each template so that the central improvement claim can be verified.","section":"Section 4"}],"minor_comments":[{"comment":"The abstract says '12.3-year' but the three sub-datasets have different livetimes: DNNC and ESTES start on 13 May 2011 while NT starts on 1 June 2010. Please state exactly which livetime is used in the combined sample and how the '12.3-year' figure is defined.","section":"Section 2"},{"comment":"The notation in Eqs. (1) and (2) is ambiguous because the index i is used for both the event index and the declination/energy bin. Please introduce separate indices for events and bins, and define \\bar D_i and \\bar S_i precisely.","section":"Section 3, Eqs. (1) and (2)"},{"comment":"Please define 'sensitivity' and 'discovery potential' directly in the table captions and state whether the 5σ threshold is local or trial-corrected. For the KRA and CRINGE rows, 'units of model flux' should be defined explicitly, and Table 2 should state whether the 'Template' columns refer to the signal hypothesis used in the likelihood when injecting the Fermi-π0 best-fit flux.","section":"Tables 1 and 2 captions"},{"comment":"The sentence 'Table 1 shows the necessary flux for 50% of pseudo experiments to exceed the TS corresponding to a 5σ discovery' appears to describe only the 'Discovery Potential' row, not the 'Sensitivity' row. Please make this explicit so that the two quantities are not conflated.","section":"Section 3, text before Table 1"},{"comment":"The figure shows the numbers of overlapping events, but the text should also state the total number of events in the combined sample after overlap removal and explain how the overlap assignment affects the signal and background PDFs.","section":"Figure 3"}],"recommendation":"major_revision","confidential_remarks":"The paper is a short ICRC proceedings contribution, and the core issue is that the main numerical claim is presented without the definitions needed to verify it. The inconsistency between Tables 1 and 2 is real and should be fixed before publication, as should the mismatch between the abstract and the body regarding the PeVatron searches. I see no fundamental flaw in the approach itself; the issues appear fixable with clarification and additional quantitative details."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"You should know this is a short ICRC proceedings describing the ICEMAN dataset — a combined all-sky sample of DNN cascades, starting tracks, and northern through-going tracks — and a template-based galactic plane sensitivity projection. The main new product is the dataset combination itself, with an explicit overlap-removal procedure (overlaps kept in the starting-track set). The likelihood (Eqs. 1–3) is standard unbinned template fitting, and the effective areas and PDF smearing are described clearly. If the projected sensitivities hold, that >20% improvement over the previous DNN-cascade analysis is a genuinely useful result.\n\nThe soft spot is not minor: Tables 1 and 2 contradict the text. Table 1 gives the Fermi-π0 discovery potential as 18.7 in units of 10^-12 TeV cm^-2 s^-1; the text says the flux measured in [3] is above the 5σ threshold. The previous measurement's best-fit flux in those same units is about 4.8, well below 18.7. If the discovery potential is correct, then injecting 4.8 should not yield 5.5σ; if the 5.5σ median is correct, then the discovery potential should be below 4.8, not four times higher. The paper never states whether the 5σ threshold in Table 1 is local or trial-corrected, which makes the comparison even less verifiable. This is load-bearing: the headline claims rest on it.\n\nOther issues are smaller. There is no systematic-uncertainty treatment; for a sensitivity projection that is common at ICRC, but it should be said explicitly. The expected significance comes from injecting the previous best-fit flux, which is a benchmark, not a measurement. The Cygnus and point-source PeVatron searches are advertised but deferred. The references include prior IceCube analyses and the relevant gamma-ray templates; nothing odd there.\n\nNet: the methodology is coherent and the dataset could be valuable, but as printed the central numbers do not add up. I would not cite the sensitivity or significance values until the authors clarify Tables 1 and 2. It deserves a referee, but mainly so the authors are forced to fix the inconsistency.","headline":"New combined IceCube sample, but the headline 5.5 sigma claim is not internally consistent with the paper's own discovery potential table.","tokens_in":10590,"tokens_out":7747,"would_cite":false,"duration_ms":84314,"reading_group":"maybe","serious_thinker":"no","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Merging three IceCube event selections into one 12.3-year all-flavor dataset lifts galactic plane template sensitivity by more than 20% and projects a 5.5 sigma detection if the previously measured flux is correct.","keywords":["galactic plane neutrinos","IceCube","all-flavor neutrino sample","ICEMAN","PeVatron search","template likelihood analysis","Cygnus region","multi-messenger astrophysics"],"falsifier":"Run the same unbinned likelihood on the real 12.3-year data set and compare the observed trial-corrected significance for the Fermi $\\pi^0$ template with the projected $5.5\\sigma$: if the real significance falls well short, or if the sensitivity computed with the three sub-samples fitted separately beats the combined sensitivity, the central claim is undermined. A sharper version: rerun the pseudo-experiments with overlap events reassigned from ESTES to Northern Tracks; if the gain vanishes, the overlap convention is carrying the improvement.","tokens_in":9538,"feed_emoji":"🌌","tokens_out":9273,"duration_ms":92267,"temperature":0.7,"pith_summary":"This paper argues that IceCube can see the Milky Way's neutrino glow more sharply by merging its three event morphologies—cascade showers, starting muon tracks, and through-going northern tracks—into one all-sky, all-flavor dataset called ICEMAN that spans 12.3 years. Using that combined sample in an unbinned template likelihood against four galactic emission models, the authors project a median local significance of $5.5\\sigma$ for the galactic plane under the flux measured by the previous cascade-only analysis, with sensitivities improved by more than 20% over that earlier work. If the projection holds, the same dataset becomes a sharper instrument for the two searches the paper announces: a template-based probe of the Cygnus Cocoon region as a galactic PeVatron, and a point-source correlation search with the more than forty gamma-ray sources above 100 TeV reported by LHAASO, HAWC, and HESS. The payoff is that neutrinos tied to known TeV gamma-ray emitters would be direct evidence that those galactic objects accelerate cosmic rays to PeV energies.","feed_headline":"Sample fusion lifts galactic plane neutrino hunt to 5.5 sigma","feed_subtitle":"Merging cascades, starting tracks, and northern tracks promises a sharper view of the Milky Way in neutrinos.","key_machinery":"The load-bearing object is the ICEMAN dataset itself: three largely independent event selections combined into one all-flavor, full-sky sample, with the largest overlap region (between ESTES and Northern Tracks) kept in the starting-track set to maximize sensitivity. The argument is carried by an unbinned maximum likelihood that models each event as a mixture of a signal term and a declination-dependent background, with the background estimated from data and corrected by signal subtraction (Eqs. 1-2), and a test statistic defined as the likelihood ratio $TS = 2\\ln[\\mathcal{L}(\\hat{n}_s)/\\mathcal{L}(n_s=0)]$. Signal templates from four galactic emission models are folded with the effective area of each sub-sample, smeared with per-event 2D Gaussian kernels matched to angular uncertainty, and weighted by the predicted energy spectrum to build spatial and energy PDFs.","core_discovery":"The central claim is that the IceCube Multi-Flavor Astrophysics Neutrino sample (ICEMAN)—the combination of 85,199 DNN-selected cascades, 11,755 starting tracks (ESTES), and the Northern Track sample, with overlapping events retained in the starting-track set—yields a template-based galactic plane measurement whose sensitivity improves by more than 20% over the previous DNN-cascade analysis. When the best-fit flux of that earlier measurement is injected into pseudo-experiments, the Fermi $\\pi^0$ template yields a median local significance of $5.5\\sigma$, with the other templates (KRA$_5^\\gamma$, KRA$_{50}^\\gamma$, CRINGE) giving $4.71\\sigma$, $4.46\\sigma$, and $5.28\\sigma$ under the same injected flux. The paper also establishes discovery potentials for all four templates and lays out two planned uses of ICEMAN: a template-based probe of the Cygnus Cocoon region and a point-source search correlating IceCube neutrinos with gamma-ray sources above 100 TeV reported by LHAASO, HAWC, and HESS.","pith_inferences":["If the $5.5\\sigma$ projection is realized, the cascade sub-sample's superior energy resolution (about 3% at 10 TeV) could let a follow-up analysis measure the spectral index of the galactic emission, which would discriminate $\\pi^0$-decay from leptonic production—a step the paper does not itself take.","A cross-check the paper leaves implicit: reassign the overlap events to the Northern Track set instead of ESTES and recompute the sensitivities; if the gain persists, the claimed improvement is robust to the overlap convention.","The stacking logic can be re-run as LHAASO, HAWC, and HESS publish more sources above 100 TeV: the ICEMAN dataset is fixed, so each new catalog entry is a new test of the same PeVatron hypothesis.","Because NT covers only the northern sky while cascades and starting tracks cover the full sky, the combined sample's southern-sky sensitivity depends almost entirely on the two event topologies with worse angular resolution, so the improvement over the previous cascade-only analysis may be smaller for southern galactic regions than the all-sky average suggests."],"forward_implications":["Under the Fermi $\\pi^0$ hypothesis the expected median local significance is $5.5\\sigma$; the other three templates (KRA$_5^\\gamma$, KRA$_{50}^\\gamma$, CRINGE) give $4.71\\sigma$, $4.46\\sigma$, and $5.28\\sigma$ under the same injected flux.","Sensitivities improve by more than 20% relative to the previous DNN-cascade galactic plane analysis for all four tested templates, making ICEMAN the default dataset for future extended galactic source searches.","The discovery potentials in Table 1 give concrete flux targets for a $5\\sigma$ claim under each model: $18.7 \\times 10^{-12}$ TeV cm$^{-2}$ s$^{-1}$ per flavor at 100 TeV for Fermi $\\pi^0$, and 0.40–0.53 in model flux units for the other three templates.","The same likelihood machinery that fits the galactic plane templates is the basis of the two announced PeVatron searches: a Cygnus Cocoon template fit and a multi-messenger-informed point-source search over gamma-ray sources above 100 TeV."],"supporting_citations":[{"why":"The previous DNN-cascade galactic plane analysis: supplies the 4.5 sigma baseline, the best-fit flux injected in pseudo-experiments, and the sensitivity the new sample must beat.","marker":"[3]"},{"why":"The 10-year starting-track (ESTES) event selection that is the basis of the starting-track sub-sample in ICEMAN.","marker":"[4]"},{"why":"The companion proceedings describing improvements to the cascade reconstruction (ice model, detector calibration) that are incorporated into the DNN cascade sample.","marker":"[7]"},{"why":"The Northern Track sub-sample, previously used in IceCube point-source searches, is the through-going track component of ICEMAN.","marker":"[8]"},{"why":"The Fermi pi-zero galactic emission template, the primary model hypothesis behind the headline significance.","marker":"[9]"},{"why":"The KRA5gamma and KRA50gamma galactic emission templates used as alternative hypotheses in the likelihood.","marker":"[10]"},{"why":"The CRINGE galactic emission template used as an alternative hypothesis in the likelihood.","marker":"[11]"},{"why":"Adds the unresolved-sources component to the CRINGE template.","marker":"[12]"}],"fun_headline_variants":["ICEMAN dataset sharpens galactic neutrino view to 5.5 sigma","Merged neutrino samples reveal galactic plane at 5.5 sigma","IceCube's ICEMAN sample boosts galactic plane sensitivity","5.5 sigma galactic plane neutrino signal from ICEMAN","New all-flavor sample homes in on galactic neutrinos"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The projected sensitivity assumes the three event samples can be folded into one unbinned likelihood with only a declination-dependent background and a shared treatment of overlapping events, and that differences in ice model, calibration, and effective area among the samples do not degrade the combined performance.","fun_headline_variants_meta":{"raw":{"variants":["ICEMAN dataset sharpens galactic neutrino view to 5.5 sigma","Merged neutrino samples reveal galactic plane at 5.5 sigma","IceCube's ICEMAN sample boosts galactic plane sensitivity","5.5 sigma galactic plane neutrino signal from ICEMAN","New all-flavor sample homes in on galactic neutrinos"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000173,"raw_usage":{"total_tokens":1337,"prompt_tokens":1059,"completion_tokens":278,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":675,"completion_tokens_details":{"reasoning_tokens":203}},"tokens_in":675,"tokens_out":278,"duration_ms":3755,"temperature":1.0,"reasoning_tokens":203,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T18:09:35.340820+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the same unbinned likelihood on the real 12.3-year data set and compare the observed trial-corrected significance for the Fermi $\\pi^0$ template with the projected $5.5\\sigma$: if the real significance falls well short, or if the sensitivity computed with the three sub-samples fitted separately beats the combined sensitivity, the central claim is undermined. A sharper version: rerun the pseudo-experiments with overlap events reassigned from ESTES to Northern Tracks; if the gain vanishes, the overlap convention is carrying the improvement.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"The companion proceedings describing improvements to the cascade reconstruction (ice model, detector calibration) that are incorporated into the DNN cascade sample."},{"cited_title":"Extending the IceCube search for neutrino point sources in the Northern sky with additional years of data","cited_arxiv_id":"2308.12742","evidence_quote":"The Northern Track sub-sample, previously used in IceCube point-source searches, is the through-going track component of ICEMAN."}],"review_version":1}