{"id":"48fa7fc4-1bfc-4978-9196-dd1429cbb60e","arxiv_id":"1908.11392","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"VEXAS DR1 cross-matches VISTA near-infrared data with WISE, DES, PanSTARRS, SkyMapper, X-ray, radio, and spectroscopic surveys to build a Southern-sky multi-wavelength catalog of 198 million objects.","lead":"This paper presents VEXAS, a new public catalog combining near-infrared data from VISTA with infrared, optical, X-ray, radio, and spectroscopic surveys for about 198 million objects in the Southern sky. It aims to be a reference dataset for classifying stars, galaxies, and quasars and for hunting rare objects such as gravitational lenses.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central superlative claim lacks a defined comparison baseline, and the catalog purity assertion relies on an unquantified AllWISE match requirement.","rationale":"The reader's weakest_assumption identifies exactly the same load-bearing concern: the 3-arcsec (or 10-arcsec) positional match is assumed to identify the same physical object, and the AllWISE requirement is assumed to eliminate spurious detections, but no false-match rate is provided. The strongest_claim (widest and deepest Southern Hemisphere multi-wavelength catalogue) rests on the completeness and purity of the cross-matched product. The paper's own text repeatedly asserts purity ('completely eliminate') without demonstration, making this an internally testable correctness risk rather than a mere difference from consensus. The superlative claim also lacks a baseline comparison; however, even a perfect match would not establish 'widest and deepest' without defined comparison surveys and metrics. I agree with the reader's CONDITIONAL verdict and with their identification of the weakest assumption. I would add a specific procedural test: analysing the angular-separation distribution of the matches, which is standard practice in cross-match papers and would directly quantify the chance-alignment contamination. The reader's verdict of CONDITIONAL is appropriate because the catalogue itself is likely useful and real, but the purity assertion is unvalidated and the headline superlative is unsupported as stated.","tokens_in":18790,"tokens_out":1678,"duration_ms":14690,"concrete_test":"For a random subsample of 10,000 objects in VEXAS-AllWISE3 (3-arcsec matches), compute the angular separation distribution and compare it against the expected distribution for true matches plus a flat background of chance alignments using the local source density. Then cross-match the same subsample against Gaia DR2 to flag objects with high proper motion or with a second Gaia source within 3 arcsec; confirm whether the VISTA and optical centroid offset is correlated with proper motion or source confusion. Report the implied false-match rate and the fraction of objects whose VISTA and optical centroids are inconsistent at the level seen in the Y-band saturation analysis of Section 4.1.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The paper's central claim (Section 8) is that VEXAS is the 'widest and deepest public optical-to-IR photometric and spectroscopic database in the Southern Hemisphere.' No comparison table, metric, or explicit baseline is provided to substantiate 'widest and deepest.' Section 2.3 even cautions that VHS-DR6 'does not match the latest release from the ESO archive.' More critically, Sections 2.3, 3, and 6.1-6.2 impose an AllWISE match within 10 arcsec and claim this requirement 'minimizes, if not completely eliminates, the number of spurious detections' (Section 1) and 'limits (or completely eliminates) spurious detections' (Section 3). No false-match rate, contamination estimate, or validation against a higher-resolution survey is provided. The 3-arcsec optical matches similarly lack an angular-separation distribution or a test for systematic offsets, such as proper motion, source confusion in dense regions, or PSF-size effects. Since the catalogue's central value is its purity and completeness, the absence of any quantitative validation of the primary cross-match assumption is the weakest load-bearing point. This is an internally testable concern, not a disagreement with consensus.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper presents VEXAS DR1, a set of nine cross-matched multi-wavelength catalogs built from VISTA (VHS DR6 and VIKING DR3) with AllWISE, DES, PanSTARRS, SkyMapper, ROSAT/XMM, SUMSS, and SDSS/6dFGS, covering the Southern Galactic Hemisphere (b < -20 deg). The authors describe the matching radii and object counts, and demonstrate two applications: selection of roughly 10^6 extremely red objects and a search for strong lens candidates using astrometric offsets. The paper claims that VEXAS is currently the widest and deepest public optical-to-IR photometric and spectroscopic database in the Southern Hemisphere.","tokens_in":19153,"tokens_out":5880,"duration_ms":52090,"significance":"If the catalog is released as described and its purity is validated, VEXAS DR1 would be a valuable community resource for Southern-sky multi-wavelength studies, enabling object classification and rare-object searches over approximately 2x10^8 objects. The paper's strengths include the public release via ESO Phase-3, the detailed specification of ADQL queries and matching radii, the use of known lenses for characterizing the lens-search technique, and the demonstration of two science applications. However, the headline superlative and the purity claims are not yet quantitatively supported.","major_comments":[{"comment":"The claim that VEXAS is the 'widest and deepest public optical-to-IR photometric and spectroscopic database in the Southern Hemisphere' lacks a defined comparison baseline. Please add a table comparing area, source density, depth limits, and wavelength coverage with existing merged catalogs (e.g., AllWISE+2MASS, DES+WISE, SkyMapper+WISE, or other public multi-wavelength products). Without such a baseline, this superlative claim cannot be evaluated.","section":"Abstract; Section 8"},{"comment":"The assertion that requiring an AllWISE match 'minimizes, if not completely eliminates, spurious detections' is not quantified. Please estimate the false-match rate for the 3\"/10\" matching radii, for example via angular-separation distributions, random radial-shift tests, or visual inspection of subsamples, and discuss how systematic astrometric offsets (proper motion, PSF-size effects, source confusion) affect the matches. This is load-bearing because catalog purity is a central claim of the paper.","section":"Sections 1, 3, and 6"},{"comment":"There is an inconsistency in the number of 21cm matches: Section 6 states '34192 have a 21cm match', while Section 6.2 reports 'This yielded 77'338 radio sources' and Table 2 lists VEXAS-21cm = 77,338. Please reconcile these counts and ensure all numbers agree between text and tables.","section":"Section 6 vs Table 2"},{"comment":"The use of VHS-DR6, which the authors note does not match the latest ESO archive release, could affect the completeness and uniformity claims. Please quantify the missing area or depth relative to the latest release, or justify explicitly why DR6 is sufficient for the stated aims of uniform coverage and depth.","section":"Section 2.3"}],"minor_comments":[{"comment":"The sentence 'The H-band from VISTA and the u-band from SM are the two least covered bands with≈ 3.2 and≈ 3.8 of objects respectively' should read 'with approximately 3.2% and approximately 3.8% of objects respectively'.","section":"Section 4.3"},{"comment":"Section 6.1 states 'thereby yielding 3044 unique sources with an X-Ray', while Section 6 reports '3049 have a match in X-Ray'; please reconcile this 5-source discrepancy.","section":"Section 6.1 vs Section 6"},{"comment":"Typographical errors: 'Karl-Schwarschild-Str.' should be 'Karl-Schwarzschild-Str.'; 'ESO Progam' should be 'ESO Program'; 'Sky Matter' should be 'SkyMapper' in Section 4.3; 'the the 6dF Galaxy Survey' has a duplicated 'the' in the Acknowledgements.","section":"Various"},{"comment":"The phrase 'comprising of 9 cross-matched' should be 'comprising 9 cross-matched'.","section":"Abstract"}],"recommendation":"major_revision","confidential_remarks":"This is a data-release paper with clear community value, but the refereeing should focus on quantitative validation of the purity and completeness claims, and on reconciling the internal count inconsistencies, before acceptance."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a useful data-release paper, and the catalog is probably a real resource. The reader's conditional verdict is fair, and the stress-test concern lands.\n\nWhat is actually new: the cross-match itself. VEXAS DR1 combines VHS and VIKING with AllWISE, then attaches DES, PanSTARRS, SkyMapper, SDSS, 6dFGS, ROSAT/XMM, and SUMSS into a single southern-sky table of roughly 198 million objects. Nothing in the cited literature offers that footprint at that depth. The paper is specific about query criteria, matching radii (10\" for infrared/X-ray/radio, 3\" for optical), and object counts at every stage. It also flags the VHS-DR6 versus ESO-archive discrepancy, which is the kind of honesty you want from a catalog paper. The two science cases show the tables have immediate utility; the ERO candidate list is large, and the known-lens recovery in the astrometric-offset section is a legitimate sanity check.\n\nWhere it is soft: the load-bearing assumption is that a positional match identifies the same physical object. The paper repeatedly says the AllWISE requirement \"completely eliminates\" spurious detections, but it never provides a false-match rate, a separation distribution, or a check against a higher-resolution survey. With the WISE PSF at about 6\" and southern-sky source densities, a 10\" match will have some contamination, and that claim needs numbers. The \"widest and deepest\" superlative also lacks a defined baseline: there is no comparison table against 2MASS+AllWISE, the ESO VHS DR4 release, or any other existing product. Section 2.3 admits a newer ESO release exists, which makes the absence of that comparison more conspicuous. Separately, the data-access link in the preprint is a placeholder, and the abstract calls the lens systems \"discoveries\" when they are candidates. These are all fixable in revision.\n\nNone of this sinks the paper. The internal numbers are consistent, the construction is transparent, and the application examples are honest demonstrations of use rather than new physical claims. The open questions about purity and completeness are exactly the kind a referee should ask.\n\nRecommendation: send this to peer review. With a quantitative contamination estimate, a clear comparison baseline for the superlative, and a working data link, it becomes a solid reference catalog paper. I would cite the catalog and would expect others to as well.","headline":"VEXAS DR1 is a genuinely useful southern-sky multi-wavelength cross-match, but the 'widest and deepest' superlative and the 'completely eliminates spurious detections' claim are both asserted without the quantitative support they need.","tokens_in":19541,"tokens_out":1833,"would_cite":true,"duration_ms":21163,"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":"VEXAS assembles nine surveys into the widest, deepest public optical-to-infrared database of the Southern Hemisphere.","keywords":["astronomical databases","catalogs","surveys","virtual observatory tools","multi-wavelength photometry","cross-matching","extremely red objects","gravitational lensing"],"falsifier":"Take a random sample of a few thousand objects from VEXAS-AllWISE3 and compare each 3-arcsecond VISTA–AllWISE pair with high-resolution imaging (Gaia or Hubble); if more than a few percent of pairs turn out to be two unrelated sources superimposed along the line of sight, the catalogue is not as clean as claimed.","tokens_in":18562,"feed_emoji":"🔭","tokens_out":7491,"duration_ms":61708,"temperature":0.7,"pith_summary":"This paper presents VEXAS-DR1, the first public release of a set of nine multi-wavelength catalogues covering the southern Galactic hemisphere, built by cross-matching the near-infrared VISTA Hemisphere Survey and VIKING catalogues with the AllWISE mid-infrared survey and then with optical, X-ray, radio, and spectroscopic data. The paper's central claim is that VEXAS is currently the widest and deepest public optical-to-infrared photometric and spectroscopic database in the Southern Hemisphere, with 198,608,360 objects each detected in at least two surveys. If that holds, VEXAS becomes a reference multi-wavelength base for classifying stars, galaxies, and quasars, computing photometric redshifts, and hunting rare objects such as strongly lensed quasars and extremely red galaxies. As demonstrations, the authors use the released tables to assemble 1,010,205 extremely red object candidates (59 with secure spectra) and to find new strong gravitational lens candidates from astrometric offsets between surveys.","feed_headline":"One catalogue, nine surveys: 198 million Southern-sky objects","feed_subtitle":"Public VEXAS tables give every object optical-to-radio matches for classification and rare-object searches.","key_machinery":"The central object is the multi-wavelength cross-matched catalogue itself, assembled by positional matching: each VISTA source is required to have an AllWISE match within 10 arcseconds (later 3 arcseconds), and the resulting table is matched to optical, X-ray, radio, and spectroscopic catalogues at fixed radii (3 arcseconds for optical, 10 arcseconds for X-ray and radio). The requirement of a match in at least two surveys, and in particular an AllWISE match, is the mechanism the paper relies on to suppress spurious detections. For the lens search, the load-bearing tool is the field-corrected astrometric offset: the centroid shift between, say, DES and VISTA is compared to the local average offset in a 0.5-degree patch, so that large residual offsets flag close blends of a quasar and a galaxy. For the extragalactic table, the WISE colour cut $W1-W2>0.2+\\sqrt{\\delta W1^2+\\delta W2^2}$ selects objects with mid-infrared excess, isolating AGN and high-redshift galaxy candidates.","core_discovery":"On its own terms, the paper establishes a data product rather than a new astrophysical result: VEXAS-DR1 provides, for every entry, photometry from at least one VISTA near-infrared band plus a match in at least one other survey, with native-system magnitudes, errors, coordinates, separations, stellarity indicators, and Galactic extinction. The core table contains 198,608,360 objects; restricting the AllWISE match to 3 arcseconds yields VEXAS-AllWISE3 with 126,372,293 objects, which is the starting point for the optical matches: 37,615,619 with DES, 24,693,386 with Pan-STARRS1, 20,331,041 with SkyMapper, 5,742,300 with extragalactic WISE colours, 347,076 with SDSS/6dFGS spectroscopy, 77,338 with SUMSS 21 cm radio, and 3,049 with X-ray detections from ROSAT/XMM. The authors claim this combination is the widest and deepest public optical-to-IR photometric and spectroscopic database in the Southern Hemisphere, and they support the claim with two applications: a statistically large sample of extremely red objects and the discovery of strong gravitational lens candidates beyond the reach of earlier searches.","pith_inferences":["A quantitative false-match rate is not given; if one were measured by matching VEXAS entries to high-resolution imaging, users would learn how much of the 198-million-object sample is truly clean rather than merely presumed clean.","The lens-candidate search via astrometric offsets could be combined with machine-learning morphology scores on the same VEXAS images to produce a much larger candidate list; the paper's examples suggest but do not quantify the completeness of the offset-only method.","Because the extremely red object selection uses only $K_s$, $i$, and $r$ cuts, the 1,010,205 candidates are a pre-selection; matching to the spectroscopic table yields only 59 confirmations, so most candidates await spectroscopic follow-up, and the true ERO fraction of the sample is an open testable question.","If the catalogue is indeed as clean as claimed, it becomes a natural training set for machine-learning classifiers separating stars, galaxies, and quasars using colours alone, since every object carries native-system magnitudes from independent surveys."],"forward_implications":["The released tables give Southern-hemisphere astronomers a ready-made reference for object classification over roughly 198 million sources, with optical, near-infrared, and mid-infrared colours in a single query.","Photometric-redshift work can use the many-band overlap (up to nine bands for a few million objects) and the 347,076 spectroscopic matches as calibration sets.","Rare-object searches—lensed quasars, extremely red objects, AGN—can be run directly on the public tables; the paper demonstrates 1,010,205 extremely red object candidates and new lens candidates found via astrometric offsets.","The extragalactic colour table isolates about 5.7 million objects with AGN-like or galaxy-like WISE colours, a large sample for studying AGN demographics.","Future data releases extending the same procedures to the northern Galactic hemisphere and to Gaia/HST matches would produce an all-sky version of the same resource."],"supporting_citations":[{"why":"Supplies the VISTA Hemisphere Survey near-infrared photometry that forms the bulk of the VEXAS input table.","marker":"[McMahon et al. 2013]"},{"why":"Supplies the VIKING near-infrared survey, the complementary footprint that adds 10.95 million objects.","marker":"[Sutherland 2012]"},{"why":"Supplies the AllWISE source catalogue, the mid-infrared match whose existence is the main spurious-detection guard.","marker":"[Cutri & et al. 2013]"},{"why":"Supplies DES optical photometry for the 37.6-million-object VEXAS-DESW table.","marker":"[Abbott et al. 2018]"},{"why":"Supplies Pan-STARRS1 optical photometry for the VEXAS-PS1W table.","marker":"[Chambers et al. 2016]"},{"why":"Supplies SkyMapper optical photometry for the VEXAS-SMW table.","marker":"[Wolf et al. 2018]"},{"why":"Supplies 6dFGS spectroscopic redshifts used in the VEXAS-SPEC table.","marker":"[Jones et al. 2004]"},{"why":"Supplies maximum-likelihood ROSAT/XMM-to-AllWISE matches that seed the VEXAS-Xray table.","marker":"[Salvato et al. 2018]"},{"why":"Supplies the SUMSS 21 cm radio catalogue used for VEXAS-21cm.","marker":"[Bock et al. 1999]"},{"why":"Defines the field-corrected astrometric-offset method used to find lensed-quasar candidates in VEXAS.","marker":"[Spiniello et al. 2018]"}],"fun_headline_variants":["VEXAS DR1: 198M objects, 9 surveys, one southern sky","Nine surveys, one catalog: VEXAS opens the southern sky","VEXAS: 198M Southern objects with optical-to-radio matches","VEXAS DR1: cross-matching nine surveys for 198M objects","VEXAS DR1: 198M multi-wavelength matches in the South"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"That a positional match within 3 arcseconds (or 10 arcseconds for X-ray and radio) between two surveys identifies the same physical object, and that requiring an AllWISE match completely eliminates spurious detections; the paper asserts this but reports no false-match or contamination rate.","fun_headline_variants_meta":{"raw":{"variants":["VEXAS DR1: 198M objects, 9 surveys, one southern sky","Nine surveys, one catalog: VEXAS opens the southern sky","VEXAS: 198M Southern objects with optical-to-radio matches","VEXAS DR1: cross-matching nine surveys for 198M objects","VEXAS DR1: 198M multi-wavelength matches in the South"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001422,"raw_usage":{"total_tokens":5855,"prompt_tokens":1180,"completion_tokens":4675,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":796,"completion_tokens_details":{"reasoning_tokens":4571}},"tokens_in":796,"tokens_out":4675,"duration_ms":32941,"temperature":1.0,"reasoning_tokens":4571,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T10:16:12.418084+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a random sample of a few thousand objects from VEXAS-AllWISE3 and compare each 3-arcsecond VISTA–AllWISE pair with high-resolution imaging (Gaia or Hubble); if more than a few percent of pairs turn out to be two unrelated sources superimposed along the line of sight, the catalogue is not as clean as claimed.","supporting_citations":[{"cited_title":"2012, Science from the Next Generation Imaging and Spectro- scopic Surveys, 40","cited_arxiv_id":null,"evidence_quote":"Supplies the VIKING near-infrared survey, the complementary footprint that adds 10.95 million objects."},{"cited_title":"M., & et al","cited_arxiv_id":null,"evidence_quote":"Supplies the AllWISE source catalogue, the mid-infrared match whose existence is the main spurious-detection guard."},{"cited_title":"A., et al","cited_arxiv_id":null,"evidence_quote":"Supplies SkyMapper optical photometry for the VEXAS-SMW table."},{"cited_title":"2018, VizieR Online Data Catalog, J/MNRAS/473/4937","cited_arxiv_id":null,"evidence_quote":"Supplies maximum-likelihood ROSAT/XMM-to-AllWISE matches that seed the VEXAS-Xray table."},{"cited_title":"C.-J., Large, M","cited_arxiv_id":null,"evidence_quote":"Supplies the SUMSS 21 cm radio catalogue used for VEXAS-21cm."},{"cited_title":"R., et al","cited_arxiv_id":null,"evidence_quote":"Defines the field-corrected astrometric-offset method used to find lensed-quasar candidates in VEXAS."}],"review_version":1}