{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:XQ72UZRDMR226KMCPQAOV3E3EH","short_pith_number":"pith:XQ72UZRD","schema_version":"1.0","canonical_sha256":"bc3faa66236475af29827c00eaec9b21d0bc68bc1e52ec0c47d196ed5e7bae26","source":{"kind":"arxiv","id":"2512.11982","version":2},"attestation_state":"computed","paper":{"title":"Semantic search for 100M+ galaxy images using AI-generated captions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.LG"],"primary_cat":"astro-ph.IM","authors_text":"Francois Lanusse, Irina Espejo, Jo Bovy, Liam Parker, Nolan Koblischke, Shirley Ho","submitted_at":"2025-12-12T19:06:14Z","abstract_excerpt":"Finding scientifically interesting phenomena through slow manual labeling campaigns severely limits our ability to explore the billions of galaxy images produced by telescopes. In this work, we develop a pipeline to create a semantic search engine from completely unlabeled image data. Our method leverages Vision-Language Models (VLMs) to generate descriptions for galaxy images, then contrastively aligns a pre-trained astronomy foundation model with these embedded descriptions to produce searchable embeddings at scale. We find that current VLMs provide descriptions that are sufficiently informa"},"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":"2512.11982","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"astro-ph.IM","submitted_at":"2025-12-12T19:06:14Z","cross_cats_sorted":["cs.AI","cs.CV","cs.LG"],"title_canon_sha256":"6163d52da342eb7b0590725907e35403bb22795248731c025f5967b019bc154b","abstract_canon_sha256":"78c1fb61bb806fad9e468668f5369b553dbc1442fd6aca8b8879de3fe746a765"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-11T01:09:24.354277Z","signature_b64":"37L5Q3CC9jXhLCyxhsFXpNWZrucf0eX7C61EpMXGdozy+qCH49JL8FbI+AQL3/kQ4I6JgnYQnvX/r8MNYHwkBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bc3faa66236475af29827c00eaec9b21d0bc68bc1e52ec0c47d196ed5e7bae26","last_reissued_at":"2026-06-11T01:09:24.353217Z","signature_status":"signed_v1","first_computed_at":"2026-06-11T01:09:24.353217Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Semantic search for 100M+ galaxy images using AI-generated captions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.LG"],"primary_cat":"astro-ph.IM","authors_text":"Francois Lanusse, Irina Espejo, Jo Bovy, Liam Parker, Nolan Koblischke, Shirley Ho","submitted_at":"2025-12-12T19:06:14Z","abstract_excerpt":"Finding scientifically interesting phenomena through slow manual labeling campaigns severely limits our ability to explore the billions of galaxy images produced by telescopes. In this work, we develop a pipeline to create a semantic search engine from completely unlabeled image data. Our method leverages Vision-Language Models (VLMs) to generate descriptions for galaxy images, then contrastively aligns a pre-trained astronomy foundation model with these embedded descriptions to produce searchable embeddings at scale. We find that current VLMs provide descriptions that are sufficiently informa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2512.11982","kind":"arxiv","version":2},"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/2512.11982/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":"2512.11982","created_at":"2026-06-11T01:09:24.353377+00:00"},{"alias_kind":"arxiv_version","alias_value":"2512.11982v2","created_at":"2026-06-11T01:09:24.353377+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2512.11982","created_at":"2026-06-11T01:09:24.353377+00:00"},{"alias_kind":"pith_short_12","alias_value":"XQ72UZRDMR22","created_at":"2026-06-11T01:09:24.353377+00:00"},{"alias_kind":"pith_short_16","alias_value":"XQ72UZRDMR226KMC","created_at":"2026-06-11T01:09:24.353377+00:00"},{"alias_kind":"pith_short_8","alias_value":"XQ72UZRD","created_at":"2026-06-11T01:09:24.353377+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2606.07779","citing_title":"Do Vision-Language Models See Dwarf Galaxies the Way We Do?","ref_index":12,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XQ72UZRDMR226KMCPQAOV3E3EH","json":"https://pith.science/pith/XQ72UZRDMR226KMCPQAOV3E3EH.json","graph_json":"https://pith.science/api/pith-number/XQ72UZRDMR226KMCPQAOV3E3EH/graph.json","events_json":"https://pith.science/api/pith-number/XQ72UZRDMR226KMCPQAOV3E3EH/events.json","paper":"https://pith.science/paper/XQ72UZRD"},"agent_actions":{"view_html":"https://pith.science/pith/XQ72UZRDMR226KMCPQAOV3E3EH","download_json":"https://pith.science/pith/XQ72UZRDMR226KMCPQAOV3E3EH.json","view_paper":"https://pith.science/paper/XQ72UZRD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2512.11982&json=true","fetch_graph":"https://pith.science/api/pith-number/XQ72UZRDMR226KMCPQAOV3E3EH/graph.json","fetch_events":"https://pith.science/api/pith-number/XQ72UZRDMR226KMCPQAOV3E3EH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XQ72UZRDMR226KMCPQAOV3E3EH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XQ72UZRDMR226KMCPQAOV3E3EH/action/storage_attestation","attest_author":"https://pith.science/pith/XQ72UZRDMR226KMCPQAOV3E3EH/action/author_attestation","sign_citation":"https://pith.science/pith/XQ72UZRDMR226KMCPQAOV3E3EH/action/citation_signature","submit_replication":"https://pith.science/pith/XQ72UZRDMR226KMCPQAOV3E3EH/action/replication_record"}},"created_at":"2026-06-11T01:09:24.353377+00:00","updated_at":"2026-06-11T01:09:24.353377+00:00"}