{"id":"0750d3e3-655d-4add-ab78-b0ae6d79935f","arxiv_id":"1908.05170","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"IceCube presents a new stacking search for neutrinos from AGN cores, with preliminary sensitivity estimates indicating that the radio-selected AGN sample alone cannot saturate the diffuse neutrino flux.","lead":"This paper describes a new IceCube search for neutrinos coming from the cores of active galaxies, using a decade of data and a new way to pick candidate galaxies by their X-ray, radio, and infrared light. It reports how sensitive the search would need to be to spot such neutrinos; no neutrinos have been found yet.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 'not saturate' conclusion is a sensitivity projection, not a measured result: Figure 5's expected flux is tied to the discovery potential, and the analysis is explicitly unblinded only in future work.","rationale":"The reader's CONDITIONAL verdict is appropriate. The paper is an honest ICRC proceedings contribution that clearly labels its results as preliminary, but the central 'not saturate' statement is more fragile than the X-ray weighting issue alone. The analysis is still blinded, and the key comparison in Figure 5 is made with discovery-potential quantities rather than with an unblinded flux measurement or a model prediction with a fixed absolute normalization. Figure 4's own wording ('would have to produce in order for a discovery to be made') shows that the curves encode detector sensitivity, not a measured or theoretically predicted neutrino flux. The Section 2 proportionality L_nu ∝ L_X provides a weighting scheme but no coefficient, so the vertical scale of the 'expected flux' curves in Figure 5 is not fixed by the pγ/pp models cited. The concrete check that would settle the concern is to unblind the stacking analysis and compute an upper limit on the neutrino flux from the radio-selected AGN sample; if that upper limit is below the diffuse flux at γ = 2, the central claim would become empirical. This does not change the reader's conditional verdict, but it sharpens the condition under which the claim should be accepted.","tokens_in":6536,"tokens_out":10221,"duration_ms":110933,"concrete_test":"Reconstruct the quantity plotted as 'expected neutrino flux' in Figure 5 from the definitions in Section 5 and Figure 2. If the light-blue curves are obtained by multiplying the integrated X-ray flux of all AGN by the discovery-potential-to-X-ray ratio, then they are sensitivity curves, not model predictions; the decisive check is to unblind the 10-year stacking analysis for the radio-selected AGN sample and compute the 90% CL upper limit on the total neutrino-flux normalization. If the resulting upper limit lies below the IceCube diffuse flux at γ = 2, the saturation claim is empirically supported; if it lies above, or if no upper limit is produced, the Section 5 statement remains a projection and should be rephrased as an anticipated sensitivity rather than a measured conclusion.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The most load-bearing concern is that the central claim in Section 5 — 'the neutrinos expected from these AGN are not going to saturate the diffuse neutrino flux' — is not supported by data yet. The paper explicitly states that the discovery-potential values in Figure 2 are preliminary, with the last two sub-samples extrapolated, and that full unblinded results are deferred to a future publication. Figure 4 defines the plotted quantity as 'what fraction of the diffuse neutrino flux the AGN would have to produce, in order for a discovery to be made'; Figure 5 then compares curves derived from this sensitivity quantity with the diffuse flux. A discovery-potential curve lying above the diffuse flux shows that a population emitting the full diffuse flux would not necessarily yield a 5σ discovery in this analysis; it does not measure, or even upper-bound, the true neutrino output of the radio-selected AGN population. The claim also inherits the unnormalized assumption of Section 2 that neutrino luminosity is proportional to soft X-ray flux; the text gives no model-derived absolute normalization from the cited pγ/pp scenarios, so the vertical placement of the 'expected flux' curves is set by detector sensitivity rather than by a neutrino-production model. Until the analysis is unblinded and an upper limit is compared with the diffuse flux, the saturation statement remains an expectation, not a result.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper, an IceCube Collaboration proceedings contribution for ICRC 2019, proposes a stacking search for high-energy neutrinos from the cores of active galactic nuclei (AGN). Three AGN samples are built from radio, infrared, and X-ray catalogues (NVSS, AllWISE, 2RXS/XMMSL2), with blazars removed, and sources are weighted by soft X-ray flux as a proxy for accretion-disk luminosity; the LLAGN sample is additionally weighted by a 'Seyfertness' probability. An unbinned maximum-likelihood ratio test is applied to ten years of IceCube data. The paper presents preliminary discovery-potential curves for the radio-selected sample and compares them with the IceCube diffuse flux, concluding that neutrinos from these AGN are 'not going to saturate' the observed diffuse flux. Unblinded results are deferred to a future publication.","tokens_in":6790,"tokens_out":4650,"duration_ms":43607,"significance":"The paper addresses an important open problem, the origin of the bulk of the IceCube astrophysical neutrino flux, and the source-selection strategy is thoughtful: cross-matching X-ray, radio, and infrared catalogues while removing gamma-ray blazars is a reasonable way to isolate AGN-core candidates. The statistical machinery is standard IceCube methodology, and the authors are transparent that the results are preliminary. If the central claim were actually measured, it would be a valuable constraint on AGN-core neutrino production. As it stands, however, the main conclusion is a sensitivity projection, not an observational result, so the paper's immediate value is as a forecast and a description of the analysis plan rather than as a measurement.","major_comments":[{"comment":"The statement that 'the neutrinos expected from these AGN are not going to saturate the diffuse neutrino flux seen by IceCube' is not supported by the plotted curves. The light-blue curves are discovery-potential curves (the flux required for a 5-sigma discovery in 50% of trials), as made explicit in Figure 2. A discovery-potential curve lying above the diffuse-flux line means that a population that emitted the full diffuse flux would not be detected at 5-sigma by this analysis; it does not measure or upper-bound the true neutrino output of the radio-selected AGN population. Because the analysis has not been unblinded, the saturation claim should be rephrased as a sensitivity statement, e.g., 'this analysis is not yet sensitive enough to detect a population that saturates the diffuse flux,' or removed entirely.","section":"Section 5 and Figure 5"},{"comment":"The label 'expected neutrino flux' in Figure 5 is misleading. No absolute normalization from the pp or p-gamma models of Section 2 is derived; the assumption L_nu is proportional to soft X-ray flux fixes only the relative weighting of sources, not the overall neutrino flux. The vertical placement of the 'expected flux' curves is therefore set by detector sensitivity (the discovery potential) and by the assumed diffuse-flux normalization, not by a neutrino-production model. The same issue affects Figure 4, where the plotted quantity is correctly described in the text as 'the fraction of the diffuse neutrino flux the AGN would have to produce in order for a discovery to be made' but could easily be misread as a measured fraction. Please relabel these curves as required fluxes or sensitivities and state explicitly that they are not predictions of the models discussed in Section 2.","section":"Section 5 and Figures 4/5"},{"comment":"The conclusion in Figure 5 uses the full radio-selected sample of 13,927 sources, but Figure 2 indicates that the discovery-potential values for the last two sub-samples, which include the full sample, are extrapolated (dashed line) because a larger number of trials is required. Thus the central comparison in Figure 5 rests on an extrapolated sensitivity value. The authors should either provide the required trials for the full sample, or clearly exclude the full-sample point from the conclusion and restrict the claim to the non-extrapolated sub-samples.","section":"Section 5 and Figure 2"}],"minor_comments":[{"comment":"The term 'Radiative Inefficient Acceleration Flows (RIAFs)' is incorrect; the standard term is 'Radiatively Inefficient Accretion Flows'.","section":"Section 2"},{"comment":"The 'Matched catalogues' row is garbled and should be cleaned up; for example, the three samples should read 'NVSS + 2RXS/XMMSL2', 'AllWISE + 2RXS/XMMSL2', and '2RXS + XMMSL2'.","section":"Table 1"},{"comment":"The phrase 'a softer neutrino spectrum, which is a better description of the IceCube diffuse flux' is vague; the specific spectral index intended should be stated.","section":"Section 5"},{"comment":"The axis label 'log10S(0.5-2 keV)' should include units consistently, e.g., 'log10[S(0.5-2 keV)/erg cm^-2 s^-1]'.","section":"Figure 3"}],"recommendation":"major_revision","confidential_remarks":"This is a preliminary proceedings paper, and the main quantitative claim is a sensitivity projection rather than a measured result. The revision path is clear: the authors need to reframe the saturation statement as a forecast, relabel the curves, and avoid relying on the extrapolated full-sample point. No deeper concerns about novelty or methodology beyond those stated in the report."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Take this as an ICRC proceedings paper, not a final result. What is new: the first IceCube stacking search explicitly targeting non-jetted AGN cores, with a careful multi-catalog construction (radio, IR, X-ray) and a Seyfertness PDF that lets you weight LLAGN candidates. The cross-matching work looks solid: they clean against 3LAC, handle duplicates, check their source counts against X-ray luminosity functions, and are upfront that the discovery-potential values for the largest stacks are extrapolated. That is real work and will be a useful reference for anyone building AGN source lists for neutrino analyses.\n\nThe soft spots are exactly where the reader's report puts them. The statement in Section 5 that 'the neutrinos expected from these AGN are not going to saturate the diffuse neutrino flux' sounds like a physical prediction, but it is actually a statement about detector sensitivity. Figure 5's 'expected flux' lines are tied to the discovery potential, not to a neutrino-production model with an absolute normalization. The pp/pγ scenarios in Section 2 are described only qualitatively; no formula maps X-ray luminosity to a neutrino flux. So the vertical placement of those curves says 'at this flux a 5σ discovery would not be achieved,' not 'the AGN population emits below the diffuse flux.' The stress-test note is right: until unblinding and an actual upper limit, the saturation claim is an expectation, not a result.\n\nTwo smaller things. First, the L_nu ∝ L_X weighting is an assumption and the paper doesn't test its sensitivity to it. That is standard in this literature, but since the whole discovery potential depends on it, a sentence on how wrong it could be would have helped. Second, the extrapolated discovery-potential points for O(10^3) sources are dashed, but they still anchor the comparison in Figure 5; readers should not treat those as measured sensitivities.\n\nNone of this makes the paper unserious. It is honest, methodologically clean, and on a genuinely important question. The citation pattern is appropriate; the IceCube self-citations are to standard references. As a proceedings paper it deserves publication and a reader can learn from the sample construction. As a claim about the AGN core origin of the diffuse flux, it is a sensitivity projection awaiting unblinding. I would send it to peer review in this form, and I would not cite it as evidence about the diffuse flux composition, only as a description of an ongoing search.","headline":"First IceCube stacking search specifically targeting non-jetted AGN cores, with a genuinely useful sample-selection method, but its 'not saturate' conclusion is a sensitivity projection, not a measured result.","tokens_in":7311,"tokens_out":3216,"would_cite":true,"duration_ms":32439,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["98.54.Cm","95.85.Ry"],"model":"deepseek-v4-flash","headline":"Searching ten years of IceCube data, this paper asks whether accretion-disk cores of AGN, not their jets, produce the observed high-energy neutrinos, and reports that the expected contribution from radio-selected AGN cores falls short of…","keywords":["AGN cores","IceCube","high-energy neutrinos","stacking analysis","accretion disk","X-ray luminosity","radio galaxies","diffuse neutrino flux"],"falsifier":"Perform the same stacking analysis with neutrino weights set by a different accretion-disk proxy, such as hard X-ray or mid-infrared luminosity, and compare the best-fit neutrino flux: if the result changes substantially, the linear X-ray scaling is falsified. Alternatively, if the unblinded best fit from a large AGN sample is consistent with zero while the model requires more than 100% of the IceCube diffuse flux at γ = 2, the AGN-core scenario is excluded.","tokens_in":6322,"feed_emoji":"🔭","tokens_out":4631,"duration_ms":43268,"temperature":0.7,"pith_summary":"This paper asks whether the high-energy neutrinos IceCube detects could come from the cores of Active Galactic Nuclei, specifically from protons accelerated near the supermassive black hole interacting with accretion-disk radiation, rather than from relativistic jets. It builds three large samples of AGN (radio-selected, infrared-selected, and low-luminosity AGN), removes blazar contamination, and stacks ten years of IceCube data with an unbinned maximum-likelihood search. Each source is weighted by its soft X-ray flux, used as a proxy for accretion-disk luminosity and therefore expected neutrino output. The preliminary result for the radio-selected sample is that the neutrino flux expected from these AGN cores does not saturate the observed diffuse neutrino flux, meaning this population alone cannot account for all of IceCube's neutrinos. The analysis is still blind, and the discovery-potential estimates suggest a detection is possible if AGN cores supply a large enough fraction of the diffuse flux.","feed_headline":"AGN cores alone can't saturate IceCube's neutrino flux","feed_subtitle":"Stacking ten years of IceCube data shows radio-selected AGN cores fall short of the diffuse neutrino flux.","key_machinery":"The central machinery is an unbinned maximum-likelihood stacking analysis adapted from IceCube's point-source searches. Each AGN enters the signal probability density function weighted by its expected relative neutrino contribution, which is taken to be proportional to the measured soft X-ray flux (0.5–2 keV) as a proxy for accretion-disk luminosity. Source samples are built by positional cross-matching of the NVSS radio catalogue, the AllWISE infrared catalogue, and the 2RXS and XMMSL2 X-ray catalogues, with blazars removed using the 3LAC catalogue; for LLAGN an additional 'Seyfertness' probability is used as a weight. To compare with the diffuse flux, the total X-ray flux of all AGN is estimated using X-ray luminosity functions, and the required fraction of the IceCube diffuse neutrino flux is computed for a γ = 2 spectrum.","core_discovery":"The central claim is quantitative: for radio-selected AGN cores (13,927 sources, weighted by 0.5–2 keV X-ray flux, and assuming a neutrino spectrum with spectral index γ = 2), the expected neutrino flux lies below the IceCube diffuse neutrino flux. Figure 5 shows the expected flux from the AGN sample (both 13,927 and 1,000 stacked sources) above the diffuse-flux line, so the paper states that the neutrinos expected from these AGN are not going to saturate the diffuse neutrino flux seen by IceCube. At the same time, the fraction of diffuse flux required for a 5σ discovery is below unity for much of the accessible stacked X-ray flux, so the search has the sensitivity to detect a sub-saturating contribution if AGN cores are responsible for a substantial part of the diffuse flux.","pith_inferences":["A null or sub-saturating result across all three AGN samples would argue that neutrino emission from accretion disks is subdominant, shifting attention to source classes whose neutrino output is not tightly correlated with X-ray luminosity, such as choked jets or star-forming galaxies.","The assumed linear X-ray-to-neutrino scaling could itself be tested by splitting the stacked sample by X-ray hardness or by black-hole-mass estimates and checking whether the neutrino flux follows the same proportionality in each subset.","The same X-ray-weighting scheme could be cross-checked with IceCube's high-energy starting events, which are less affected by atmospheric-muon backgrounds than muon-track samples.","If a positive stacking signal emerges after unblinding, the per-source energy-flux/X-ray-flux ratio measured here would provide a direct calibration for future multi-messenger searches targeting AGN cores."],"forward_implications":["If AGN cores produce neutrinos in proportion to their X-ray flux, then radio-selected AGN cores alone contribute less than the full IceCube diffuse flux at γ = 2, so other source populations or a different production mechanism are needed to fill the gap.","The analysis sensitivity is about three times better than the 5σ discovery potential, so the final unblinded search can constrain the AGN-core contribution to a level below the diffuse flux.","Applying the same stacking procedure to the IR-selected AGN and LLAGN samples, and to softer neutrino spectra, is expected to yield tighter limits and will further test the accretion-disk origin of the diffuse neutrino flux."],"supporting_citations":[{"why":"Provides the Shakura-Sunyaev thin-disk model whose 'big blue bump' photons serve as targets for pγ neutrino production.","marker":"[10]"},{"why":"Supplies the RIAF proton-acceleration and neutrino-production model used for the LLAGN scenario.","marker":"[12]"},{"why":"Provides the AllWISE-matched 2RXS and XMMSL2 X-ray catalogues used as the basis for source selection.","marker":"[16]"},{"why":"Defines the ten-year IceCube dataset used in the stacking analysis.","marker":"[21]"},{"why":"Describes the unbinned maximum-likelihood ratio test used to compute the stacking significance.","marker":"[22]"},{"why":"Provides one of the X-ray luminosity functions used to estimate the total X-ray flux of all AGN for the diffuse-flux comparison.","marker":"[23]"},{"why":"Provides the X-ray luminosity function used for the LLAGN total flux estimate.","marker":"[25]"},{"why":"Supplies the IceCube diffuse muon-neutrino flux with γ = 2 used as the normalization and comparison line in Figures 4 and 5.","marker":"[26]"}],"fun_headline_variants":["AGN cores underdeliver for IceCube's neutrino budget","Radio-selected AGN cores fall short of neutrino flux","AGN cores can't explain IceCube's diffuse neutrinos","Stacked AGN search finds neutrino flux shortfall","IceCube data show AGN cores not saturating flux"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The analysis assumes that each AGN's neutrino luminosity is proportional to its soft X-ray flux; if neutrino output depends on other factors such as black hole spin, disk state, or jet orientation, the stacking weights and the comparison to the diffuse flux would not be valid.","fun_headline_variants_meta":{"raw":{"variants":["AGN cores underdeliver for IceCube's neutrino budget","Radio-selected AGN cores fall short of neutrino flux","AGN cores can't explain IceCube's diffuse neutrinos","Stacked AGN search finds neutrino flux shortfall","IceCube data show AGN cores not saturating flux"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000201,"raw_usage":{"total_tokens":1382,"prompt_tokens":950,"completion_tokens":432,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":566,"completion_tokens_details":{"reasoning_tokens":349}},"tokens_in":566,"tokens_out":432,"duration_ms":4750,"temperature":1.0,"reasoning_tokens":349,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:21:01.826818+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Perform the same stacking analysis with neutrino weights set by a different accretion-disk proxy, such as hard X-ray or mid-infrared luminosity, and compare the best-fit neutrino flux: if the result changes substantially, the linear X-ray scaling is falsified. Alternatively, if the unblinded best fit from a large AGN sample is consistent with zero while the model requires more than 100% of the IceCube diffuse flux at γ = 2, the AGN-core scenario is excluded.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the Shakura-Sunyaev thin-disk model whose 'big blue bump' photons serve as targets for pγ neutrino production."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the RIAF proton-acceleration and neutrino-production model used for the LLAGN scenario."},{"cited_title":"Salvato et al., Monthly Notices of the Royal Astronomical Society 473 (2017) 4937–4955","cited_arxiv_id":null,"evidence_quote":"Provides the AllWISE-matched 2RXS and XMMSL2 X-ray catalogues used as the basis for source selection."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Describes the unbinned maximum-likelihood ratio test used to compute the stacking significance."},{"cited_title":"Miyaji, G","cited_arxiv_id":null,"evidence_quote":"Provides one of the X-ray luminosity functions used to estimate the total X-ray flux of all AGN for the diffuse-flux comparison."},{"cited_title":"Hasinger, T","cited_arxiv_id":null,"evidence_quote":"Provides the X-ray luminosity function used for the LLAGN total flux estimate."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the IceCube diffuse muon-neutrino flux with γ = 2 used as the normalization and comparison line in Figures 4 and 5."}],"review_version":1}