{"id":"d65eb196-7de5-438a-ae35-d92f6462be87","arxiv_id":"2606.04208","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"HETDEX PDR1 releases an expanded public catalog of 426,654 LAEs, 491,411 [O II] emitters and additional sources from 86.67 sq deg of integral-field spectroscopy with improved classifiers.","lead":"The paper releases the first public data from the HETDEX survey, a large integral-field spectroscopic map covering 86.67 square degrees that catalogs over 426,000 Lyman-alpha emitting galaxies plus other sources. Smart generalists should read it because the public catalog and data cubes give direct access to a major new dataset for studying high-redshift galaxies and cosmic expansion.","discovery_kind":"extension","skeptic_critique":{"model":"grok-4.3","headline":"Classifier validation metrics (purity, completeness, contamination rates) are not reported, so the precise class counts rest on an unverified assumption.","rationale":"The reader's weakest assumption is exactly the load-bearing step. The paper's central deliverable is the classified catalog; without quantified validation the counts cannot be taken at face value. A CONDITIONAL verdict reflects that the data release itself is valuable but the class-specific numbers require the missing validation evidence to be accepted without caveat.","tokens_in":1956,"tokens_out":302,"duration_ms":13147,"concrete_test":"Locate the methods or appendix section describing the ML classifiers; extract any reported accuracy, F1 scores, confusion matrix, or estimated contamination fraction per class on a held-out sample. If those metrics are absent or show contamination >5% for the dominant classes, recompute the headline counts after applying a conservative purity correction and compare to the published numbers.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The headline claim states exact counts (426654 LAEs, 491411 [O II], etc.) obtained after applying new ML classifiers and improved QC. These counts are only meaningful if the classifiers achieve low contamination and the classes are cleanly separated. The abstract describes the classifiers and QC but supplies no performance numbers, training details, test-set results, or contamination estimates. Because the reported numbers are the direct output of that step, any undetected misclassification directly scales the catalog contents.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.3","summary":"The manuscript presents HETDEX Public Data Release 1 (PDR1), consisting of 431,713 IFU observations over 86.67 sq deg and the HETDEX Public Source Catalog 2 (HPSC2). HPSC2 reports 426,654 LAEs, 491,411 [O II] emitters, 19,457 low-z galaxies, 18,303 AGN, and 150,608 stars, each with coordinates, redshifts/velocities, and 1D spectra, derived from new ML classifiers and improved QC on four additional years of data. Data cubes, a 1.6M-candidate LAE sample, and raw databases are also released via the HETDEX portal and Zenodo.","tokens_in":2046,"tokens_out":355,"duration_ms":12464,"significance":"If the source classifications hold, PDR1 supplies one of the largest public samples of high-redshift LAEs for large-scale structure and dark-energy studies at cosmic noon, together with supporting [O II] and stellar samples. Public access to cubes, spectra, and a JupyterLab environment directly supports community re-use and reproducibility.","major_comments":[{"comment":"Abstract and HPSC2 processing description: the headline source counts (426,654 LAEs etc.) are the direct output of the new ML classifiers and QC pipeline, yet no purity, completeness, contamination fractions, training-set sizes, or test-set performance metrics are supplied. Without these numbers the reliability of the class separations cannot be assessed and the catalog contents remain unverifiable.","section":"Abstract / HPSC2 description"}],"minor_comments":[],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":{"model":"grok-4.3","summary":"We thank the referee for their careful review of the manuscript. We address the single major comment below and will make the requested additions to improve the verifiability of the catalog.","responses":[{"response":"We agree that quantitative performance metrics are necessary to allow readers to assess the reliability of the ML-based classifications in HPSC2. In the revised manuscript we will add a dedicated subsection (placed after the description of the new classifiers) that reports: (i) sizes of the training and validation sets used for each classifier, (ii) test-set performance metrics including precision, recall, F1-score, and confusion-matrix summaries obtained via cross-validation, and (iii) estimated purity, completeness, and contamination fractions for the five source classes. These values will be presented both in the text and in a summary table. The added material will be drawn from the internal validation analyses already performed during classifier development and will be placed in the main text rather than an appendix so that the headline counts can be interpreted in context.","revision_made":"yes","referee_comment":"[Abstract / HPSC2 description] Abstract and HPSC2 processing description: the headline source counts (426,654 LAEs etc.) are the direct output of the new ML classifiers and QC pipeline, yet no purity, completeness, contamination fractions, training-set sizes, or test-set performance metrics are supplied. Without these numbers the reliability of the class separations cannot be assessed and the catalog contents remain unverifiable."}],"tokens_in":1581,"tokens_out":322,"duration_ms":13894,"standing_objections":[]},"desk_editor":{"model":"grok-4.3","letter":"The main takeaway is that HETDEX has put out an expanded public source catalog from their integral-field survey, now covering 86.67 sq deg with 431k observations. It incorporates extra data since the 2023 release, better quality control, and fresh machine learning classifiers to tag LAEs, [O II] emitters, low-z galaxies, AGN, and stars. The catalog gives coordinates, redshifts or velocities, and 1D spectra for each object, all available through their portal and Zenodo.\n\nWhat works is the straightforward delivery of the data products. They include the full detection catalog, a 1.6 million candidate LAE sample, and the cubes themselves. The survey design stays consistent with prior HETDEX papers, which keeps the release usable for large-scale structure work at 1.88 < z < 3.52.\n\nThe soft spot is the lack of any reported numbers on the new classifiers. The abstract lists exact counts—426k LAEs, 491k [O II], etc.—but supplies no purity, completeness, contamination rates, or test-set results. Those numbers are the direct output of the classification step, so without those metrics it is hard to gauge how much misclassification might be baked in. The data cubes also carry the usual caveat that local sky subtraction makes them unsuitable for surface brightness work.\n\nThis paper is aimed at anyone who needs a large, public sample of high-redshift emission-line galaxies for cosmology or galaxy evolution studies. It is the kind of release that deserves a serious referee to check the processing pipeline and validation details rather than a desk reject.","headline":"This is a standard data release paper that adds four years of HETDEX observations and new classifiers to produce a larger public catalog, but the class counts depend on unshown classifier performance.","tokens_in":2710,"tokens_out":413,"would_cite":true,"duration_ms":13498,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.3","headline":"HETDEX Public Data Release 1 catalogs 426,654 LAEs plus 491,411 OII emitters with positions, redshifts and spectra","keywords":["HETDEX","Lyman-alpha emitters","emission-line galaxies","integral-field spectroscopy","public data release","source catalog","large-scale structure"],"falsifier":"Independent spectroscopic follow-up of a random subsample from each class that finds contamination rates substantially above those implied by the catalog would falsify the separation.","tokens_in":2858,"feed_emoji":"🔭","tokens_out":712,"duration_ms":19475,"temperature":0.7,"pith_summary":"The paper presents the first public data release from the HETDEX integral-field spectroscopic survey of the sky. HPSC2 expands prior catalogs with four more years of data, new machine learning classifiers, and better quality control to classify and deliver coordinates, redshifts or velocities, and 1D spectra for 426,654 Lyman-alpha emitters at 1.88 < z < 3.52, 491,411 OII emitters, and additional low-redshift galaxies, AGN, and stars. The release covers 86.67 square degrees of non-contiguous fields plus legacy regions and supplies the associated data cubes. A sympathetic reader cares because the untargeted sample supplies ready-to-use data for mapping large-scale structure at cosmic noon without new telescope time.","feed_headline":"HETDEX releases catalog of 1.1 million sources over 87 sq deg","feed_subtitle":"Public catalog gives positions, redshifts and spectra for 426k LAEs and 491k OII emitters to map structure at cosmic noon","key_machinery":"HPSC2, the reprocessed source catalog that applies machine learning classifiers and quality control to integral-field spectra to separate and characterize emission-line sources.","core_discovery":"HPSC2 contains 426,654 LAEs, 491,411 [O II] emitters, 19,457 low-z galaxies, 18,303 active galactic nuclei, and 150,608 stars, providing coordinates, redshifts or stellar velocities, and 1D spectra for each source. The catalog derives from 431,713 IFU observations covering 86.67 sq deg and incorporates improved quality control plus new machine learning classifiers on four additional years of data.","pith_inferences":["The scale of the LAE sample may support tighter constraints on the growth of structure once clustering measurements are performed.","Raw detection databases released with the catalog allow external teams to test alternative classification methods.","The non-contiguous sky coverage suggests that future contiguous extensions could improve measurements of very large-scale modes."],"forward_implications":["The catalog supplies ready positions and spectra for statistical studies of LAE clustering over large volumes at 1.88 < z < 3.52.","1D spectra enable direct measurement of line properties and velocities for each classified object.","Inclusion of legacy fields such as COSMOS allows direct cross-matching with multi-wavelength data from other surveys.","Public data cubes support searches for faint emission lines, though local sky subtraction limits use for absolute surface brightness or very extended sources."],"fun_headline_variants":["HETDEX PDR1 catalogs 1.1M sources over 87 sq deg","HETDEX catalog lists 426k LAEs and 491k OII emitters","1.1M sources from HETDEX 87 sq deg survey released","HETDEX HPSC2 catalogs sources across 87 sq deg"],"cache_read_input_tokens":2112,"weakest_assumption_plain":"The machine learning classifiers and quality control procedures correctly separate the different source classes with low contamination.","fun_headline_variants_meta":{"raw":{"variants":["HETDEX PDR1 catalogs 1.1M sources over 87 sq deg","HETDEX catalog lists 426k LAEs and 491k OII emitters","1.1M sources from HETDEX 87 sq deg survey released","HETDEX HPSC2 catalogs sources across 87 sq deg"]},"model":"grok-4.3","cost_usd":0.011715,"raw_usage":{"total_tokens":5250,"prompt_tokens":913,"num_sources_used":0,"completion_tokens":83,"cost_in_usd_ticks":117149500,"prompt_tokens_details":{"text_tokens":913,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":4254,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":913,"tokens_out":83,"duration_ms":24790,"temperature":1.0,"reasoning_tokens":4254,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-06-28T08:54:14.300279+00:00","model_set":{"reader":"grok-4.3"},"falsifier":"Independent spectroscopic follow-up of a random subsample from each class that finds contamination rates substantially above those implied by the catalog would falsify the separation.","supporting_citations":[],"review_version":1}