{"id":"bab7a728-667d-4529-8eda-f4a9f7af05f4","arxiv_id":"2507.07895","paper_version":1,"verdict":"REJECT","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"high","formal_verification":"none","parameter_count":5,"one_line_summary":"Assuming GW190425 was a black hole merger from dark-matter-induced neutron star collapse, the authors derive dark matter constraints and forecast that only Einstein Telescope/Cosmic Explorer can confidently classify such events.","lead":"The authors interpret the ambiguous gravitational-wave event GW190425 as a possible black hole merger formed after dark matter collapsed its neutron star progenitors, and use that assumption to constrain dark matter mass and cross-section. They simulate the event under future detector networks, finding that only next-generation observatories can reliably tell a neutron star merger from a black hole merger.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The classification forecast relies on a Savage–Dickey ratio computed at a boundary where the prior density of the effective tidal deformability vanishes; the reported HLI#/DD2 ratio of 1.04 is numerically incompatible with that row's own 90% interval excluding zero, so the A# stiff-EoS…","rationale":"The reader's rejection identified both an unproven assumption about GW190425's nature and numerical inconsistencies in the classification metrics. The assumption is explicitly conditional in the paper ('we proceed under the assumption that the source of GW190425 was a BBH merger'), so it is a modeling choice rather than a hidden flaw; the more load-bearing problem is the numerical inconsistency in the classification evidence. In particular, the HLI# DD2 row of Table IV is self-contradictory: the reported posterior credible interval excludes zero while the Savage–Dickey ratio is essentially unity. Since that row is the key evidence for the abstract's statement that A# can classify stiff-EoS BNSs, the central forecasting claim is not supported by the manuscript as written. The paper does have genuine strengths: it uses public LVK posterior samples, standard Bayesian tools, and a clearly stated conditional framework, and the CE4020ET conclusions may survive a corrected analysis. However, the A# classification conclusion and the quantitative interpretation of the Savage–Dickey ratios must be redone before the paper's claims can be accepted. I therefore recommend conditional acceptance only after the classification metric is recomputed correctly and shown to be consistent with the displayed posteriors.","tokens_in":20803,"tokens_out":8289,"duration_ms":108171,"concrete_test":"Regenerate the HLI# DD2 injection with the same bilby/dynesty settings and compute the posterior density at Lambda_tilde = 0 using a KDE or sample-count in a small bin, along with the actual prior density under the component-Lambda priors and Eq. (4). If the 90% credible interval again excludes zero, a valid point-density Savage–Dickey ratio must be much smaller than unity; a ratio near 1.04 indicates a computational or definitional error in the metric. Independently, compute the Bayes factor between the BBH and BNS models from nested-sampling evidences for this injection; this avoids the boundary issue and would settle whether HLI# can actually classify a stiff-EoS BNS.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The paper's abstract and Sec. V claim that an A# (HLI#) network can confidently classify a GW190425-like BNS provided the equation of state is stiff. That claim is directly undercut by the paper's own Table IV. For the HLI# DD2 injection, the reported effective tidal deformability is Lambda_tilde = 311.8 +193.1 -199.1, so the 90% credible interval is approximately [112.7, 504.9] and excludes Lambda_tilde = 0. Yet the same row reports a Savage–Dickey ratio of 1.04, which would require the posterior density at zero to be comparable to the prior density at zero. A posterior whose credible interval excludes zero cannot have such a density ratio. Moreover, the Savage–Dickey ratio is not well defined here: with uniform priors on the component Lambdas in Eq. (4), the induced prior density of Lambda_tilde vanishes at Lambda_tilde = 0, so the denominator of the ratio is zero. Appendix A's prescription of evaluating the ratio 'on a grid of points with Lambda < 1' replaces a point density with an arbitrary interval probability, so the reported numbers are not Savage–Dickey ratios in the required sense. Thus the supporting evidence for the HLI# stiff-EoS classification result is internally inconsistent, and the A# conclusion in the abstract and conclusions is not established by the analysis as presented.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reanalyzes GW190425 under the hypothesis that it was a binary black hole (BBH) formed by dark-matter-induced collapse of neutron stars, using the LVK posterior for the effective tidal deformability. It then simulates GW190425-like BNS and BBH injections for future detector networks (HLVK+, HLI#, CE4020ET) and uses Savage–Dickey ratios and posterior widths to forecast source classification. Under the BBH assumption, the paper uses Fisher-matrix forecasts and collapse-time formulas from Singh et al. to derive constraints on the dark-matter mass and cross-section, both for O3 and for future networks. The abstract's main claims are that an A#-sensitivity network can classify a stiff-EoS BNS event, that next-generation observatories can classify even soft-EoS events, and that a single low-mass BBH detection can constrain asymmetric dark matter.","tokens_in":21076,"tokens_out":9824,"duration_ms":110911,"significance":"If the analysis were correct, the paper would offer a novel way to constrain asymmetric dark matter from a single low-mass merger and a useful forecast for third-generation detectors. Strengths include the use of public GWTC-2.1 posterior samples, a transparent simulation setup with bilby/dynesty, and explicit statements of caveats (low SNR, model dependence of the collapse timescale, and the conditional nature of the BBH assumption). However, the classification metric is not valid as implemented, and the numerical results in Table IV are internally inconsistent. These problems affect the paper's central classification claims and, indirectly, the DM forecasts, so the significance of the results as presented is not established.","major_comments":[{"comment":"The quantity called the Savage–Dickey ratio is not actually a Savage–Dickey ratio. With uniform priors on the component tidal deformabilities Λ1 and Λ2 in Eq. (4), the induced prior density of Λ̃ vanishes at Λ̃=0, so the denominator π(Λ̃=0) in the quoted formula is zero and the ratio is undefined. The Appendix's prescription of evaluating the ratio on a grid of points with Λ̃<1 replaces a point density with an interval probability, which is not a Bayes factor for the null hypothesis Λ̃=0. Consequently, the conversion of the reported values to 'probability of a BBH origin' via B/(1+B) is not justified, and every classification probability derived from Table IV is unsupported.","section":"Appendix A / Table IV"},{"comment":"The reported Savage–Dickey ratio of 1.04 (≈51% probability of BBH origin) is inconsistent with the same row's 90% credible interval for Λ̃, [112.7, 504.9], which excludes Λ̃=0. A posterior whose 90% interval excludes zero has negligible density at zero, so a density ratio near unity is impossible; at best the interval-based quantity is not the point-null Bayes factor it is claimed to be. This directly undercuts the abstract and Sec. V statement that an A# network can confidently classify a stiff-EoS BNS event, because the table indicates a coin-flip classification for the DD2 injection.","section":"Table IV, HLI#/DD2 row"},{"comment":"The text in Sec. IVA states that for the CE4020ET APR4 injection the probability of a BBH origin is about 19%, which would require a Savage–Dickey ratio of approximately 0.23. Table IV lists a ratio of 5.2, which under the same conversion corresponds to about 84%. This is a direct numerical contradiction in the central case used to support the claim that CE4020ET provides robust classification even for the softest equation of state. The authors should either report the actual posterior probability or correct the table; as it stands, the claim is not established.","section":"Table IV, CE4020ET/APR4 row"},{"comment":"The inference of the collapse time uses the 95th percentile of the Λ̃ posterior of the same event to set the threshold σ_Λ̃T and then assumes NBBH=1 for that event. This makes the inferred collapse time depend on the event's own noise realization and posterior width, rather than on an independently calibrated population criterion. The resulting DM constraints in Figs. 8 and 9 inherit this dependence without a demonstrated calibration. A concrete test would be to repeat the inference with σ_Λ̃T fixed by an independent population threshold or by an ensemble of simulated noise realizations.","section":"Sec. IVC / Fig. 7"}],"minor_comments":[{"comment":"The axis labels in Figs. 1 and 4 contain a placeholder symbol '×10□3'; the superscripts should be restored so the figures are readable.","section":"Table IV / Fig. 4"},{"comment":"The text says that in the HLVK+ network the SNR is similar to that of GW190425, but Table IV lists SNR ≈ 23 for HLVK+ injections versus ≈ 13 for the original event; these values should be reconciled.","section":"Sec. IVA"},{"comment":"Table V reports Bayes factors ln BF_BBH/BNS but does not state which injected signal type (BBH or BNS) or equation of state was used; without this information the table is not reproducible.","section":"Table V"},{"comment":"There are sentence fragments and missing words in Sec. II, for example 'the recovery of tidal information. Due to their increased sensitivity'; the text should be revised for clarity.","section":"Sec. II / Sec. IIA"},{"comment":"Equations (9) and (10) are quoted from Ref. [28] without derivation. Because these formulas are load-bearing for the DM constraints and Ref. [28] is closely connected to the present work, a brief derivation sketch or independent check would strengthen the paper.","section":"Eqs. (9)-(10)"}],"recommendation":"major_revision","confidential_remarks":"I agree with the stress-test concern: it lands directly on Table IV and Appendix A. The classification metric is invalid as implemented, and the HLI#/DD2 and CE4020ET/APR4 rows are numerically impossible as reported. The problems are fixable in principle by recomputing a proper model comparison, but the current manuscript's central claims should not be accepted without that reanalysis. I also note that the collapse-time formulas in Eqs. (9)-(10) are taken from Ref. [28] without independent verification; given the central role of these formulas in the DM constraints, an independent derivation or validation should be requested."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper's real contribution is a concrete case study: taking the dark-matter-implosion channel from Singh et al. and asking what GW190425 would look like if it were a BBH from that channel, then forecasting how future networks would separate that from a BNS. That is genuinely new, and the authors are honest about the speculative starting point. They also do the heavy lifting of simulating injections across A+, A#, and ET/CE networks and working through the rate and dark-matter constraints. The limitations section is unusually candid.\n\nBut the classification metric they lean on is not sound as used. The Savage-Dickey ratio comparing Lambda=0 against a continuous prior is undefined here: with uniform priors on the component Lambdas, the induced prior density of Lambda_tilde is zero at Lambda_tilde=0, so the denominator is zero. Replacing the point density with an interval probability over Lambda<1 changes the quantity being reported. The table also self-contradicts: the HLI# DD2 row gives a 90% credible interval for Lambda_tilde that excludes zero (roughly 113 to 505) yet reports a Savage-Dickey ratio of 1.04, which would require the posterior and prior densities at zero to be nearly equal. Those two statements can't both be true. And the translation of ratios to probabilities is off elsewhere: the text says the CE4020ET APR4 BNS has about 19% BBH probability, but a ratio of 5.2 would give roughly 84% under the stated formula.\n\nThis is a load-bearing problem because the abstract and conclusions sell the A# stiff-EoS result on the strength of this metric. The Fisher-based forecasts and the dark-matter constraints are less affected, since they use the width of the tidal posterior rather than the ratio, but the central classification claims are not established as presented. A referee should ask for a re-analysis with a well-defined model comparison—say, a prior with a point mass at Lambda=0, or a Bayes factor that doesn't require a density at a point where the prior vanishes.\n\nI'd send this to review, not desk reject. The question is interesting, the work is substantial, and the flaw is fixable. But as it stands, the headline classification results don't hold up. Readers working on GW classification or dark-matter searches with future detectors may still find the forecast framework worth a look, but they should not take the A# conclusion at face value.","headline":"The case study is new and worth examining, but the Savage-Dickey metric is broken at Lambda=0 and the A# classification claims don't follow from the paper's own numbers.","tokens_in":21646,"tokens_out":4213,"would_cite":false,"duration_ms":44964,"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 paper argues that GW190425, often classified as a neutron-star merger, may instead be a black-hole merger from dark-matter-induced collapse, and that tidal deformability can settle the classification and constrain dark matter.","keywords":["gravitational waves","GW190425","binary neutron star","binary black hole","tidal deformability","asymmetric dark matter","dark matter-induced collapse","next-generation gravitational wave detectors"],"falsifier":"A future high-signal-to-noise observation of a GW190425-like event whose effective tidal deformability \\(\\tilde{\\Lambda}\\) is measured with a 90% credible interval excluding zero would show the source was a neutron-star binary rather than a black-hole binary, breaking the dark-matter constraint derived under the black-hole assumption. A population study finding low-mass black-hole mergers in the 1-3 \\(M_\\odot\\) range at a rate inconsistent with the predicted number from the collapse-time model would likewise indicate the formation channel or its rate normalization is wrong.","tokens_in":20554,"feed_emoji":"🌌","tokens_out":10455,"duration_ms":98807,"temperature":0.7,"pith_summary":"GW190425 is one of only two known neutron-star merger candidates, but it had no electromagnetic counterpart and its low signal-to-noise ratio leaves its nature unresolved. This paper takes seriously the possibility that it was instead a binary black hole merger, formed when asymmetric bosonic dark matter accumulated in two neutron stars and collapsed each into a black hole. Under that assumption, the event's nearly zero effective tidal deformability becomes a measurement that constrains the dark matter particle mass and its interaction cross-section with ordinary matter. The paper then simulates GW190425-like signals for future detector generations and shows that near-term upgrades cannot confidently classify such events, while a next-generation network can distinguish neutron-star from black-hole binaries using tidal information alone. If the assumption is right, a single low-mass merger could probe dark matter physics that direct-detection experiments struggle to reach.","feed_headline":"Tidal fingerprints can settle whether GW190425 was a black hole merger","feed_subtitle":"If dark matter collapsed the neutron stars first, one low-mass event could constrain its mass and cross-section.","key_machinery":"The load-bearing observable is the effective tidal deformability \\(\\tilde{\\Lambda}\\), the mass-ratio-weighted combination of the two stars' dimensionless tidal deformabilities \\(\\Lambda_1\\) and \\(\\Lambda_2\\), which is zero for black holes and strictly positive for neutron stars and enters the gravitational-wave phase at fifth post-Newtonian order. Because a black-hole signal analyzed with a neutron-star waveform returns a posterior for \\(\\tilde{\\Lambda}\\) concentrated at zero, the sharpness of that posterior, quantified by the Savage-Dickey ratio and the 90% width \\(\\sigma_{\\tilde{\\Lambda}}\\), serves as the classification metric. On the dark matter side, the central mechanism is the collapse-time inference: a merger-rate model built from the star-formation rate, a delay-time distribution, and the local binary merger rate gives the expected number of observable black-hole mergers as a function of the neutron-star-to-black-hole collapse time \\(t_c\\), and that time is then mapped to dark matter mass and cross-section through the Bose-Einstein condensation and direct-collapse formation timescales.","core_discovery":"On its own terms, the paper claims that a compact-binary gravitational-wave event with no electromagnetic counterpart and a posterior for the effective tidal deformability \\(\\tilde{\\Lambda}\\) that peaks at zero can be interpreted as a binary black hole produced by dark-matter-induced collapse of neutron stars, and that this interpretation converts the event into a dark matter constraint. The empirical hook is GW190425's \\(\\tilde{\\Lambda}\\) posterior from the GWTC-2.1 catalog, which favors zero but is broad; the authors proceed under the assumption that the source was a black-hole merger. From that assumption they derive limits on the dark matter particle mass \\(m_\\chi\\) and interaction cross-section \\(\\sigma_\\chi\\) by way of the inferred collapse time. They then forecast, using simulated injections analyzed with Bayesian inference and a Fisher-matrix population study, that a network with A+ sensitivity cannot confidently classify a GW190425-like event, that an A# network can classify it only for stiff equations of state, and that a next-generation network combining the Einstein Telescope and Cosmic Explorer separates even soft-equation-of-state neutron-star signals from black-hole signals, enabling confident classification and tighter dark matter bounds.","pith_inferences":["If the dark-matter-implosion channel is real, the apparent mass gap between neutron stars and stellar black holes should be filled with black holes near 1-3 \\(M_\\odot\\), and future population studies could look for this excess rather than automatically classifying every such event as a neutron-star binary.","The forecasting results imply that the neutron-star equation of state is now the key unknown for classification strategy: a stiff equation of state would let A# networks settle these cases, while a soft one pushes the decision to next-generation observatories.","The same Savage-Dickey-based classification metric could be applied to any low-mass merger without an electromagnetic counterpart, avoiding the Occam-penalty that makes standard Bayes factors favor the black-hole model even when tides are present.","A future high-signal-to-noise event with decisively positive \\(\\tilde{\\Lambda}\\) would falsify the dark-matter-implosion interpretation for that event, while the derived dark-matter bounds would still stand as conservative limits for events whose tidal posteriors remain consistent with zero."],"forward_implications":["If GW190425 was a dark-matter-induced black-hole merger, its observed tidal posteriors already place limits on the asymmetric dark matter mass and cross-section, complementing direct-detection bounds.","An A+-sensitivity network cannot distinguish a GW190425-like neutron-star binary from a black-hole binary, leaving classification ambiguous in the near term.","An A# network resolves the classification only if neutron stars follow a relatively stiff equation of state; for soft equations of state, the neutron-star and black-hole tidal posteriors overlap.","A next-generation Einstein Telescope plus Cosmic Explorer network distinguishes even soft-equation-of-state neutron-star mergers from black-hole mergers, because the neutron-star \\(\\tilde{\\Lambda}\\) posterior moves clearly away from zero.","A single confidently classified low-mass black-hole merger in a next-generation network would infer a much tighter collapse time than currently possible, with the remaining uncertainty dominated by the local merger-rate measurement."],"supporting_citations":[{"why":"Discovery paper for GW190425, supplying the event's masses, chirp mass, and low-signal-to-noise context.","marker":"[14]"},{"why":"Establishes the dark-matter-induced neutron-star implosion mechanism and the collapse-time formulas that map gravitational-wave observations to dark matter constraints.","marker":"[28]"},{"why":"GWTC-2.1 catalog, the source of the GW190425 posterior samples used for the effective tidal deformability inference.","marker":"[47]"},{"why":"IMRPhenomXAS waveform, the black-hole template and the base model for the neutron-star injections.","marker":"[40]"},{"why":"IMRPhenomXAS_NRTidalv3 waveform, adds the tidal phase corrections used to simulate binary neutron-star signals.","marker":"[41]"},{"why":"GWTC-2.1 confident data release, providing the median source parameters used for the simulated GW190425-like injections.","marker":"[67]"},{"why":"GWTC-3 population inference, setting the local binary merger rate that calibrates the predicted black-hole merger rate.","marker":"[74]"},{"why":"Star-formation rate used to integrate the merger rate density over redshift in the collapse-time calculation.","marker":"[75]"},{"why":"Supplies the Fisher-matrix calculation of signal-to-noise ratio and tidal-deformability uncertainty for the population forecast.","marker":"[77]"}],"fun_headline_variants":["GW190425: Tidal fingerprints could reveal its true nature","Dark matter collapse: New way to classify GW190425","Can gravity waves tell if GW190425 was a black hole?","Gravitational waves may settle GW190425's identity and dark matter"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that GW190425 really was a binary black hole merger formed when dark matter collapsed two neutron stars; the paper's evidence is a tidal-deformability posterior that peaks at zero but is too broad at the event's low signal-to-noise ratio to exclude a neutron-star binary with a soft equation of state, a limitation the authors acknowledge.","fun_headline_variants_meta":{"raw":{"variants":["GW190425: Tidal fingerprints could reveal its true nature","Dark matter collapse: New way to classify GW190425","Can gravity waves tell if GW190425 was a black hole?","Gravitational waves may settle GW190425's identity and dark matter"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000211,"raw_usage":{"total_tokens":1448,"prompt_tokens":1015,"completion_tokens":433,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":631,"completion_tokens_details":{"reasoning_tokens":360}},"tokens_in":631,"tokens_out":433,"duration_ms":5095,"temperature":1.0,"reasoning_tokens":360,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T18:31:09.518644+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A future high-signal-to-noise observation of a GW190425-like event whose effective tidal deformability \\(\\tilde{\\Lambda}\\) is measured with a 90% credible interval excluding zero would show the source was a neutron-star binary rather than a black-hole binary, breaking the dark-matter constraint derived under the black-hole assumption. A population study finding low-mass black-hole mergers in the 1-3 \\(M_\\odot\\) range at a rate inconsistent with the predicted number from the collapse-time model would likewise indicate the formation channel or its rate normalization is wrong.","supporting_citations":[{"cited_title":"Müther, M","cited_arxiv_id":null,"evidence_quote":"GWTC-3 population inference, setting the local binary merger rate that calibrates the predicted black-hole merger rate."}],"review_version":1}