{"id":"c027a3ab-7a63-46db-9ea1-64d69d5ae9c1","arxiv_id":"2509.10221","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":5.0,"correctness_risk":"low","formal_verification":"none","parameter_count":2,"one_line_summary":"High-mass X-ray binaries with Be companions predominantly move slower than 40 km/s, while those with supergiant companions predominantly move faster, a statistically robust separation likely reflecting different natal kicks.","lead":"Astronomers used Gaia data to measure the space velocities of 63 high-mass X-ray binaries and found that the two main classes, Be systems and supergiant systems, move at very different speeds. This strengthens the idea that the supernova that created each compact object left a distinguishing kick and hints that velocity can help classify these binaries.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 4/pi isotropic correction is load-bearing because 22/34 BeXRBs, but only 6/16 SgXRBs, lack direct Vr; the full-sample K-S rejection is not shown to survive on measured-Vr systems alone.","rationale":"The reader's weakest assumption is the same one I would flag, and the counts make it sharper. The K-S claim is the paper's headline result, and it combines 28 directly measured Vpec values with 35 values obtained by multiplying a 2D sky velocity by 4/pi. The correction is unbiased only under exact isotropy, and the paper's own validation (Section 6.2, Table 4) shows p=0.06 in the UV plane, just outside conventional rejection. Because the Be class has 22 of its 34 members in the isotropic subset while the Sg class has only 6 of 16, any anisotropy that correlates with line of sight will preferentially reshape the BeXRB CDF and could either create or enhance the apparent 40 km/s separation. The statement that the difference persists with measured-Vr systems is not quantified as a K-S rejection fraction, so the central claim's robustness to this assumption is currently unverified. I do not recommend rejection: the restricted-subset histograms, the MC uncertainty treatment, and the Gaia-based distances are genuine evidence, and the population-synthesis explanation is clearly labelled preliminary. But the paper should either report the restricted K-S test or soften the 'irrespective' language. The proposed check is cheap and directly settles whether the isotropy assumption is load-bearing.","tokens_in":40696,"tokens_out":16080,"duration_ms":139701,"concrete_test":"Using the paper's 50,000-sample Monte Carlo posteriors, re-run the 1,000-ensemble K-S test on the 22 systems with a measured systemic Vr only (12 BeXRBs, 10 SgXRBs, as flagged in Table 2), drawing one value per system per ensemble and applying ks_2samp with the same threshold. Report the rejection fraction and the distribution of p-values. If the rejection fraction is materially below 100 per cent, the full-sample 40 km/s segregation is dependent on the isotropic 4/pi estimates; if it remains 100 per cent, the isotropy concern is resolved. Repeat the restricted test excluding RX J1826.2-1450 and Cyg X-1 to check outlier sensitivity.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim rests on the 4/pi isotropic conversion in Section 4 applied to 35 systems lacking a literature systemic radial velocity. The correction is class-imbalanced: 22 of 34 BeXRBs but only 6 of 16 SgXRBs enter the K-S test via V_iso; the BeXRB half of the comparison is therefore anchored by the isotropy ansatz. The validation in Section 6.2 is marginal: in the UV plane the K-S p-value is 0.06 (Table 4), and that is the plane closest to the 2D sky projection for disc sources. The paper says the difference 'persists' for the 28 measured-Vr systems (Fig. 3) but does not report a K-S rejection fraction for that restricted sample, so the headline 100 per cent rejection is not shown to be independent of the isotropic estimates. If the restricted-sample rejection is also 100 per cent, the concern is answered; if not, the claim is conditional on isotropy.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper compiles 63 Galactic HMXBs with Gaia DR3 astrometry, applies zero-point-corrected parallaxes and distance inversion, and combines proper motions with literature systemic radial velocities for 28 systems to compute 3D peculiar velocities. For the 35 systems without measured Vr, it assumes isotropic peculiar motions and multiplies the 2D sky velocity by 4/pi. The authors report mean Vpec of about 21 km/s for BeXRBs and 58 km/s for SgXRBs, and an ensemble Kolmogorov-Smirnov test that rejects a common parent distribution in 100% of 1,000 Monte Carlo draws, with an optimal separation threshold near 40 km/s. They further use COSMIC/BSE population synthesis to argue that SgXRBs have shorter pre-supernova orbital periods and higher fractional mass loss than BeXRBs, and propose Vpec as a complementary classifier for HMXBs.","tokens_in":40884,"tokens_out":5946,"duration_ms":48218,"significance":"If the kinematic segregation is robust, this is a valuable three-dimensional confirmation and extension of the earlier Hipparcos-based result (Chevalier & Ilovaisky 1998), and it provides an observational probe of natal kicks in HMXBs. The paper is careful in several respects: parallax zero-point corrections, Monte Carlo propagation of astrometric and Galactic-constant uncertainties, an ensemble K-S procedure rather than a single-point comparison, and explicit sample-selection checks via colour-magnitude diagrams and K-S tests against the XRBcats parent sample. The population-synthesis link is suggestive but is presented as preliminary and needs strengthening before it can carry the causal interpretation.","major_comments":[{"comment":"The headline K-S rejection in Section 5 is not shown to be independent of the isotropic 4/pi correction, because the correction is applied unevenly: 22 of 34 BeXRBs but only 6 of 16 SgXRBs enter the comparison through V_iso. The validation of the isotropy ansatz in Section 6.2 is only marginal in the UV plane (K-S p = 0.06, Table 4), which is the plane most relevant to the sky projection used for disc sources. Figure 3 shows the restricted measured-Vr subset only qualitatively, and no K-S rejection fraction is reported for that subset. Please report the ensemble K-S result restricted to the measured-Vr Be/Sg systems, and quantify how the rejection fraction changes if the V_iso systems are excluded or if their Vpec is modelled with an anisotropic velocity distribution calibrated to the UV-plane discrepancy.","section":"Section 4; Section 6.2, Table 4"},{"comment":"The abstract claims rejection 'irrespective of the background stellar velocity dispersion', but no quantitative test varying the assumed background dispersion is presented. Section 5 only cites a typical 20 km/s young-disk dispersion, and Section 7(iii) states that accounting for background scatter 'is expected to moderate' the estimates without demonstration. Because the paper itself notes in Section 4 that true space velocities require knowledge of birth sites, the robustness claim needs a concrete test, for example convolving the Vpec distributions with Gaussians of increasing dispersion and tracking the K-S rejection fraction, or a justification that all sample members belong to young populations with known small dispersions.","section":"Abstract; Section 5; Section 7(iii)"},{"comment":"The population-synthesis interpretation is load-bearing for the causal claim in the title, but as presented it is preliminary: the simulations are described as part of upcoming work (Dashwood Brown et al., in preparation), the matching criteria and kick prescriptions are only summarized, and Figure 10 shows separated means (4.0 vs 172.5 days; 0.5 vs 0.3 fractional mass loss) without uncertainties or significance tests. Please either provide the quantitative support, including the simulated Vpec distributions and tests against the observed ones, or explicitly reframe these results as a tentative explanation rather than a demonstrated origin of the kinematic segregation.","section":"Section 6.3, Fig. 10"}],"minor_comments":[{"comment":"Section 7(iii) gives mean Vpec values of 20.2 and 48.9 km/s for BeXRBs and SgXRBs, while Section 5 reports 20.9 ± 3.5 and 58.0 ± 6.6 km/s; these numbers should be reconciled.","section":"Section 7(iii) vs. Section 5"},{"comment":"Summary item (i) says a search was conducted within a 0.5 arcsec radius, but Section 2 describes a 5 arcsec search radius; the inconsistency should be corrected.","section":"Section 7(i) vs. Section 2"},{"comment":"There are several typographical errors: 'Kinemetic' in the Section 6.3 heading, 'segration' in summary item (ix), and 'margenta' in the Fig. 14 caption.","section":"Section 6.3; Section 7(ix); Fig. 14"},{"comment":"The statement that Figs. 3 and 4 'clearly demonstrate a significant difference in the mean velocities' overstates what histograms alone show; the ensemble K-S test in the following paragraph is the appropriate statistical evidence and should be cited as such.","section":"Section 5, Figs. 3 and 4"},{"comment":"Two radial velocity values are listed for 4U 2206+543 (-62.7 and -54.5 ± 1.0 km/s) without stating which one was adopted in the kinematic analysis; please specify the adopted value and reference.","section":"Table 2, 4U 2206+543"}],"recommendation":"major_revision","confidential_remarks":"The observational analysis is generally careful and the reported segregation is plausible, but the class-imbalanced use of the 4/pi isotropic correction, together with the marginal UV-plane validation, is a genuine risk to the central claim. If the authors provide the restricted-sample K-S rejection fraction and a quantitative treatment of background velocity dispersion, I would likely support acceptance. The population-synthesis section should either be strengthened with quantitative comparisons or clearly labelled as preliminary, since the title invokes a causal natal-kick interpretation."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThis is a careful, incremental study that largely confirms what Chevalier & Ilovaisky (1998) and Fortin et al. (2022) already hinted at: BeXRBs and SgXRBs have different peculiar velocity distributions. The sample is larger (63 HMXBs), the astrometry is properly zero-point corrected, the Monte Carlo uncertainty propagation is thorough, and the 1,000-ensemble K-S test is a nice way to fold in individual posterior shapes. The paper also does a genuinely useful job on completeness and selection effects, comparing against the parent catalogue with K-S tests on magnitude and colour. The population synthesis part is openly preliminary and not load-bearing; it is used to suggest an explanation, not to prove the claim.\n\nThe main soft spot is the isotropy correction. 35 of 63 systems lack a measured systemic radial velocity, and the 4/pi multiplicative factor is applied to 22 of 34 BeXRBs but only 6 of 16 SgXRBs. So the BeXRB half of the comparison leans heavily on the isotropy ansatz. The validation in Section 6.2 is marginal for the UV plane (K-S p = 0.06), which is the plane closest to the sky projection for disc sources, and the paper does not report a K-S rejection fraction for the 28 systems with measured Vr alone. It says the difference 'persists' but never quantifies it. That is the gap the stress-test note flags, and it is a real one. The abstract's 'irrespective of the background stellar velocity dispersion' is asserted without a numerical test; Section 7.1(iii) only says the correction 'is expected to moderate' the estimates. Both are easy to fix with a sensitivity test on the restricted sample and a dispersion-adjustment check.\n\nThe 40 km/s separation threshold is chosen on the same sample used to demonstrate the difference, so calling it a classification feature is slightly optimistic. The authors do hedge this, but a validation on an independent sample or a bootstrap would strengthen it.\n\nNone of this breaks the core result. The full-sample 100% rejection is robust to Monte Carlo resampling, and the distributions look separated even by eye in the measured-Vr subset. This paper deserves a serious referee and should be publishable after the authors add the missing restricted-sample statistic and either test the background-dispersion claim or soften it.\n\nIf you work on HMXB kinematics or natal kick constraints, cite it for the sample and the method. I would bring it to a reading group to discuss the isotropy issue and the threshold selection.\n\nBest,\n[Your name]","headline":"Solid incremental confirmation of the Be/Sg velocity segregation, but the headline significance test rests on an under-tested isotropy correction and the abstract overstates robustness.","tokens_in":41439,"tokens_out":5171,"would_cite":true,"duration_ms":40835,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Supernova kicks split X-ray binaries at 40 km/s","keywords":["high-mass X-ray binaries","natal kicks","peculiar velocities","Gaia DR3","Be/X-ray binaries","supergiant X-ray binaries","binary population synthesis","Corbet diagram"],"falsifier":"Measure full three-dimensional peculiar velocities for the 35 HMXBs currently lacking systemic radial velocities, for example with time-series spectroscopy of their companion stars, and re-run the K-S test; if the Be and supergiant distributions no longer separate near 40 km/s, the segregation is an artifact of the isotropy assumption.","tokens_in":40516,"feed_emoji":"💫","tokens_out":13266,"duration_ms":105677,"temperature":0.7,"pith_summary":"The paper tests whether the supernova that forms the compact object in a high-mass X-ray binary leaves a measurable imprint on the system's space velocity, and claims that it does. Using Gaia DR3 astrometry for 63 systems, it finds that Be/X-ray binaries mostly move slower than about 40 km/s relative to Galactic rotation, while supergiant X-ray binaries mostly move faster. A Kolmogorov-Smirnov test rejects a common parent distribution in all 1,000 Monte Carlo ensembles, and the split persists when only systems with measured radial velocities are used. If right, peculiar velocity becomes a complementary fingerprint for classifying HMXBs and a new observational constraint on natal kick models.","feed_headline":"Supernova kicks split X-ray binaries at 40 km/s","feed_subtitle":"Be-star binaries drift below 40 km/s; supergiant binaries outrun it, betraying different supernova birth histories.","key_machinery":"The argument turns on the peculiar velocity $V_{\\rm pec}$, the system's three-dimensional motion after removing solar motion and circular Galactic rotation, computed from Gaia DR3 parallaxes and proper motions together with literature systemic radial velocities. For the 35 systems without radial velocities, the paper assumes isotropic peculiar motions and converts the two-dimensional sky velocity to a three-dimensional speed with the factor $4/\\pi$, the expected ratio of 3D to 2D speeds under isotropy. The 40 km/s threshold is the velocity at which the two classes' cumulative distributions separate most cleanly. Binary population synthesis supplies the physical interpretation by matching simulated binaries to observed properties and inferring pre-supernova orbital periods and fractional mass losses.","core_discovery":"The central discovery is a clean kinematic segregation in the peculiar velocities of Galactic high-mass X-ray binaries: BeXRBs cluster below about 40 km/s, SgXRBs above, and the two cumulative distributions are maximally separated near that threshold. A K-S test rejects a common parent distribution in 100 per cent of 1,000 Monte Carlo ensembles, and the difference remains when only the 28 systems with measured systemic radial velocities are considered. Binary population synthesis with the cosmic code shows SgXRBs preferentially come from pre-supernova binaries with shorter orbital periods (mean about 4 days) and higher fractional mass loss (mean about 0.5) than BeXRBs (about 172 days and 0.3, respectively), matching the prediction of van den Heuvel et al. (2000). These findings extend the Hipparcos-era two-dimensional result to full three-dimensional kinematics.","pith_inferences":["If faster SgXRBs travel farther from their birth sites, the model predicts a larger Galactic scale height and wider spatial spread for SgXRBs than BeXRBs; this spatial prediction is not made in the paper but should be testable with the same sample.","The 40 km/s split might also classify HMXBs in external galaxies where only proper motions or radial velocities are available, provided the same kick physics applies.","If future radial-velocity campaigns confirm the isotropy assumption, the $4/\\pi$ correction could be applied to much larger HMXB samples, turning proper-motion surveys into three-dimensional kinematic censuses.","The fast BeXRB outlier LS 5039, noted by the paper as a possible triple, suggests dynamical interactions can mimic or exceed natal kicks; checking multiplicity among high-peculiar-velocity BeXRBs would test that channel."],"forward_implications":["Peculiar velocity can serve as a complementary classification feature: unclassified HMXBs moving slower than about 40 km/s are likely BeXRBs, while faster systems are likely SgXRBs.","The class difference supports a natal-kick origin: SgXRBs' tighter pre-supernova orbits and higher fractional mass loss naturally produce larger systemic velocities.","The result validates using Gaia astrometry plus the isotropic $4/\\pi$ correction to recover three-dimensional kinematics when radial velocities are missing.","Population synthesis models of HMXBs should reproduce short pre-supernova periods and high mass loss for supergiant progenitors, adding a new matching constraint.","The fast BeXRB outlier RX J1826.2–1450/LS 5039 shows the 40 km/s split is statistical rather than absolute, flagging special evolutionary paths."],"supporting_citations":[{"why":"supplies the DR3 astrometry from which parallaxes and proper motions are drawn.","marker":"Gaia Collaboration 2023"},{"why":"the earlier Hipparcos-based 2D kinematic study whose class difference is extended to 3D.","marker":"Chevalier & Ilovaisky 1998"},{"why":"supplies the progenitor hypothesis of tighter pre-SN orbits and higher mass loss for supergiants.","marker":"van den Heuvel et al. 2000"},{"why":"introduces the isotropic assumption and the 4/pi conversion factor used to estimate 3D velocities.","marker":"Hobbs et al. 2005"},{"why":"provides prior 3D kinematics of 35 neutron-star HMXBs and the tendency this paper quantifies.","marker":"Fortin et al. 2022"},{"why":"gives the kinematic formulation linking orbital velocity to runaway velocity.","marker":"Kalogera 1996"},{"why":"provides the cosmic binary population synthesis code used to infer pre-SN orbital periods and mass loss.","marker":"Breivik et al. 2020"},{"why":"supplies the delayed supernova mechanism adopted in the population synthesis models.","marker":"Fryer et al. 2012"}],"fun_headline_variants":["Natal kicks split X-ray binaries at 40 km/s","40 km/s gap reveals two X-ray binary families","Be stars slow, supergiants fast: kick segregation","X-ray binaries show natal kicks segregate by type"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The result hinges on the assumption that HMXB peculiar motions are isotropic, which lets the authors turn two-dimensional sky speeds into three-dimensional space speeds for 35 of 63 systems; if the unmeasured motions are actually directional, those velocities would be biased and the 40 km/s class split could be an artifact.","fun_headline_variants_meta":{"raw":{"variants":["Natal kicks split X-ray binaries at 40 km/s","40 km/s gap reveals two X-ray binary families","Be stars slow, supergiants fast: kick segregation","X-ray binaries show natal kicks segregate by type"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000682,"raw_usage":{"total_tokens":3171,"prompt_tokens":1097,"completion_tokens":2074,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":713,"completion_tokens_details":{"reasoning_tokens":2008}},"tokens_in":713,"tokens_out":2074,"duration_ms":15251,"temperature":1.0,"reasoning_tokens":2008,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T15:56:39.383975+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure full three-dimensional peculiar velocities for the 35 HMXBs currently lacking systemic radial velocities, for example with time-series spectroscopy of their companion stars, and re-run the K-S test; if the Be and supergiant distributions no longer separate near 40 km/s, the segregation is an artifact of the isotropy assumption.","supporting_citations":[],"review_version":1}