{"paper":{"title":"Disk-like galaxies at 4 < z < 7.7 : JWST/NIRCam morphologies revealed by denoising VAE-GCNN classification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"A VAE-GCNN pipeline on JWST/NIRCam images measures a 0.34 disk-like fraction among 100 galaxies at redshifts 4 to 7.7.","cross_cats":[],"primary_cat":"astro-ph.GA","authors_text":"A. Avagyan, S.S. Mirzoyan, V.G. Gurzadyan","submitted_at":"2026-04-16T04:08:00Z","abstract_excerpt":"Understanding the prevalence of disk-like galaxies at very high redshifts is crucial for constraining the early formation of angular momentum-supported structures. The advent of JWST now permits rest-frame UV and optical morphological studies deep into cosmic epochs where disks have traditionally been considered uncommon. We apply an identical denoising VAE-GCNN classification pipeline to multi-filter JWST/NIRCam cutouts in order to obtain homogeneous, morphology-based disk fractions across the sample. Our approach comprises two steps: (i) a U-Net Variational Autoencoder (VAE) is trained to re"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"We determine the fraction of disk-like galaxies as 0.34 for a sample of JWST 100 galaxies over the redshift range 4 < z < 7.7, also in dependence on the galaxy mass range.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The denoising VAE preserves intrinsic galaxy morphology without introducing or removing disk-like features, and the GCNN classifier trained on the denoised cutouts generalizes accurately to high-redshift JWST data without significant bias from training set or redshift effects.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A denoising VAE-GCNN pipeline applied to JWST/NIRCam images yields a 0.34 fraction of disk-like galaxies at 4 < z < 7.7.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A VAE-GCNN pipeline on JWST/NIRCam images measures a 0.34 disk-like fraction among 100 galaxies at redshifts 4 to 7.7.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"9241eb4eb506a8313e122195d3df7ed0ed6c7f98f0e544b218eecc654908832f"},"source":{"id":"2604.14599","kind":"arxiv","version":1},"verdict":{"id":"83b69a5f-2482-418b-864c-fd51120deb03","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T10:47:10.308734Z","strongest_claim":"We determine the fraction of disk-like galaxies as 0.34 for a sample of JWST 100 galaxies over the redshift range 4 < z < 7.7, also in dependence on the galaxy mass range.","one_line_summary":"A denoising VAE-GCNN pipeline applied to JWST/NIRCam images yields a 0.34 fraction of disk-like galaxies at 4 < z < 7.7.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The denoising VAE preserves intrinsic galaxy morphology without introducing or removing disk-like features, and the GCNN classifier trained on the denoised cutouts generalizes accurately to high-redshift JWST data without significant bias from training set or redshift effects.","pith_extraction_headline":"A VAE-GCNN pipeline on JWST/NIRCam images measures a 0.34 disk-like fraction among 100 galaxies at redshifts 4 to 7.7."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.14599/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":23,"sample":[{"doi":"","year":2025,"title":"L., Teyssier, R., & Dekel, A","work_id":"aff246b4-a766-4abb-bb82-b105647a7d3b","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2011,"title":"2011, A&A, 532, A74","work_id":"a1aab639-8e77-4eaf-b1be-f41a1cd012c2","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"et al, 2025, ApJ, 994, 126 Capozziello S., Di Valentino E., Gurzadyan V.G., 2025, Eur","work_id":"28563a33-a64d-4d96-aaee-381b1411426d","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2026,"title":"arXiv e-prints , keywords =","work_id":"ca073fe8-ce40-4101-8cf5-4296c13d13f4","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2016,"title":"Group Equivariant Convolutional Networks","work_id":"473fade8-b16b-45a0-b764-68a0f3957588","ref_index":5,"cited_arxiv_id":"1602.07576","is_internal_anchor":false}],"resolved_work":23,"snapshot_sha256":"d8bcf242f439b01263de6da95fb2145f61199451cd8ef4d592ad784bfddb7ded","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}