{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:DORP77DDZASVYUNPMDK6YSUCND","short_pith_number":"pith:DORP77DD","schema_version":"1.0","canonical_sha256":"1ba2fffc63c8255c51af60d5ec4a8268edd1c2d5062aa3711fe20812dad17297","source":{"kind":"arxiv","id":"2405.14930","version":1},"attestation_state":"computed","paper":{"title":"AstroPT: Scaling Large Observation Models for Astronomy","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["astro-ph.GA","cs.LG"],"primary_cat":"astro-ph.IM","authors_text":"Eirini Angeloudi, Marc Huertas-Company, Michael J. Smith, Ryan J. Roberts","submitted_at":"2024-05-23T18:00:00Z","abstract_excerpt":"This work presents AstroPT, an autoregressive pretrained transformer developed with astronomical use-cases in mind. The AstroPT models presented here have been pretrained on 8.6 million $512 \\times 512$ pixel $grz$-band galaxy postage stamp observations from the DESI Legacy Survey DR8. We train a selection of foundation models of increasing size from 1 million to 2.1 billion parameters, and find that AstroPT follows a similar saturating log-log scaling law to textual models. We also find that the models' performances on downstream tasks as measured by linear probing improves with model size up"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2405.14930","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"astro-ph.IM","submitted_at":"2024-05-23T18:00:00Z","cross_cats_sorted":["astro-ph.GA","cs.LG"],"title_canon_sha256":"9dd842047465ce1d030b37f054a1246177d9385d67a1426e2d40909eb34a9735","abstract_canon_sha256":"e05aa8380e9a5a61f009ca6810505cac8ed4376419a1d892aa59668080661ce6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:22:42.887765Z","signature_b64":"gzWGrmdlB8i9Qua7ZLuEBRZZcaHoLAr5Qd7EzNezZXPpz1QGyVEanXY5/YzpQNln0PjbjTKSji5W5DoOYWF4CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1ba2fffc63c8255c51af60d5ec4a8268edd1c2d5062aa3711fe20812dad17297","last_reissued_at":"2026-07-05T08:22:42.887224Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:22:42.887224Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"AstroPT: Scaling Large Observation Models for Astronomy","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["astro-ph.GA","cs.LG"],"primary_cat":"astro-ph.IM","authors_text":"Eirini Angeloudi, Marc Huertas-Company, Michael J. Smith, Ryan J. Roberts","submitted_at":"2024-05-23T18:00:00Z","abstract_excerpt":"This work presents AstroPT, an autoregressive pretrained transformer developed with astronomical use-cases in mind. The AstroPT models presented here have been pretrained on 8.6 million $512 \\times 512$ pixel $grz$-band galaxy postage stamp observations from the DESI Legacy Survey DR8. We train a selection of foundation models of increasing size from 1 million to 2.1 billion parameters, and find that AstroPT follows a similar saturating log-log scaling law to textual models. We also find that the models' performances on downstream tasks as measured by linear probing improves with model size up"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.14930","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2405.14930/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":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2405.14930","created_at":"2026-07-05T08:22:42.887295+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.14930v1","created_at":"2026-07-05T08:22:42.887295+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.14930","created_at":"2026-07-05T08:22:42.887295+00:00"},{"alias_kind":"pith_short_12","alias_value":"DORP77DDZASV","created_at":"2026-07-05T08:22:42.887295+00:00"},{"alias_kind":"pith_short_16","alias_value":"DORP77DDZASVYUNP","created_at":"2026-07-05T08:22:42.887295+00:00"},{"alias_kind":"pith_short_8","alias_value":"DORP77DD","created_at":"2026-07-05T08:22:42.887295+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.05285","citing_title":"Toward decision-aware AI for LSST-scale time-domain astronomy","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2606.25610","citing_title":"The Galaxy's Guide to the Tokenizer: A Benchmark for Scientific Foundation Models","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2602.10207","citing_title":"Optimizing Deep Learning Photometric Redshifts for the Roman Space Telescope with HST/CANDELS","ref_index":181,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DORP77DDZASVYUNPMDK6YSUCND","json":"https://pith.science/pith/DORP77DDZASVYUNPMDK6YSUCND.json","graph_json":"https://pith.science/api/pith-number/DORP77DDZASVYUNPMDK6YSUCND/graph.json","events_json":"https://pith.science/api/pith-number/DORP77DDZASVYUNPMDK6YSUCND/events.json","paper":"https://pith.science/paper/DORP77DD"},"agent_actions":{"view_html":"https://pith.science/pith/DORP77DDZASVYUNPMDK6YSUCND","download_json":"https://pith.science/pith/DORP77DDZASVYUNPMDK6YSUCND.json","view_paper":"https://pith.science/paper/DORP77DD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.14930&json=true","fetch_graph":"https://pith.science/api/pith-number/DORP77DDZASVYUNPMDK6YSUCND/graph.json","fetch_events":"https://pith.science/api/pith-number/DORP77DDZASVYUNPMDK6YSUCND/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DORP77DDZASVYUNPMDK6YSUCND/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DORP77DDZASVYUNPMDK6YSUCND/action/storage_attestation","attest_author":"https://pith.science/pith/DORP77DDZASVYUNPMDK6YSUCND/action/author_attestation","sign_citation":"https://pith.science/pith/DORP77DDZASVYUNPMDK6YSUCND/action/citation_signature","submit_replication":"https://pith.science/pith/DORP77DDZASVYUNPMDK6YSUCND/action/replication_record"}},"created_at":"2026-07-05T08:22:42.887295+00:00","updated_at":"2026-07-05T08:22:42.887295+00:00"}