{"paper":{"title":"Variational views for self-supervised learning in radio astronomy","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"A beta-VAE supplies generative views that improve self-supervised pre-training for radio galaxy morphology when combined with standard augmentations.","cross_cats":["astro-ph.GA"],"primary_cat":"astro-ph.IM","authors_text":"Anna M. M. Scaife, Johnny Joseph Alphonse","submitted_at":"2026-02-21T18:29:31Z","abstract_excerpt":"Modern astronomical surveys are producing progressively larger and more complex datasets, making traditional supervised approaches that rely on extensive labelled catalogues increasingly difficult. Consequently, pre-training using self-supervised learning (SSL), which offers a scalable route by extracting structure directly from unlabelled images, is becoming attractive for many downstream applications. In this work we consider the use of coupled self-supervised representation learning approaches for radio galaxy morphology pre-training. In order to account for the more nuanced variations in r"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Our experiments show that combining these generative views with standard image augmentations improves downstream classification performance.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the beta-VAE reconstructions supply non-redundant morphological variations that standard augmentations miss and that these variations remain useful when transferred to the downstream classifier.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Beta-VAE reconstructions used as generative augmentations improve downstream radio galaxy classification when combined with standard image augmentations in self-supervised pre-training.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A beta-VAE supplies generative views that improve self-supervised pre-training for radio galaxy morphology when combined with standard augmentations.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"c6ef4cfcdb718d482df208d700e6fcbb924616de34c92ed011e4fcf0b88e73dd"},"source":{"id":"2602.18923","kind":"arxiv","version":2},"verdict":{"id":"bde36e69-bc00-45fb-9248-a0304696f78d","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T20:10:39.868009Z","strongest_claim":"Our experiments show that combining these generative views with standard image augmentations improves downstream classification performance.","one_line_summary":"Beta-VAE reconstructions used as generative augmentations improve downstream radio galaxy classification when combined with standard image augmentations in self-supervised pre-training.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the beta-VAE reconstructions supply non-redundant morphological variations that standard augmentations miss and that these variations remain useful when transferred to the downstream classifier.","pith_extraction_headline":"A beta-VAE supplies generative views that improve self-supervised pre-training for radio galaxy morphology when combined with standard augmentations."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2602.18923/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"041e2fbb55e5d74e096c6b1e5a7fdfcbf86f39d5e47a4228523ba48fdbeaefd6"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}