{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:TQNX5JZL74Q5FQGCST6WD4H32E","short_pith_number":"pith:TQNX5JZL","schema_version":"1.0","canonical_sha256":"9c1b7ea72bff21d2c0c294fd61f0fbd11bbe8db7a12bd123e3fffd34df3ba51b","source":{"kind":"arxiv","id":"2306.09683","version":3},"attestation_state":"computed","paper":{"title":"Scaling Open-Vocabulary Object Detection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Alexey Gritsenko, Matthias Minderer, Neil Houlsby","submitted_at":"2023-06-16T08:27:46Z","abstract_excerpt":"Open-vocabulary object detection has benefited greatly from pretrained vision-language models, but is still limited by the amount of available detection training data. While detection training data can be expanded by using Web image-text pairs as weak supervision, this has not been done at scales comparable to image-level pretraining. Here, we scale up detection data with self-training, which uses an existing detector to generate pseudo-box annotations on image-text pairs. Major challenges in scaling self-training are the choice of label space, pseudo-annotation filtering, and training efficie"},"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":"2306.09683","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-06-16T08:27:46Z","cross_cats_sorted":[],"title_canon_sha256":"acb466e4dc0f105ffe5e510c1e2b1f5db71b1a9739d13a6258634b571725d972","abstract_canon_sha256":"05bba33d644838e5223909b188b0f1ab780a3d873bc52af183caf9fae0949fba"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:21:39.407868Z","signature_b64":"jUu+tHZ027QtfjCdrSZ3otRnEmpxyZe/2d23jBnVCdPSBJ1Aw+d8TN3U1pT5XX3SAgz/MBSppmxy3KyZMMstBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9c1b7ea72bff21d2c0c294fd61f0fbd11bbe8db7a12bd123e3fffd34df3ba51b","last_reissued_at":"2026-07-05T08:21:39.407404Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:21:39.407404Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Scaling Open-Vocabulary Object Detection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Alexey Gritsenko, Matthias Minderer, Neil Houlsby","submitted_at":"2023-06-16T08:27:46Z","abstract_excerpt":"Open-vocabulary object detection has benefited greatly from pretrained vision-language models, but is still limited by the amount of available detection training data. While detection training data can be expanded by using Web image-text pairs as weak supervision, this has not been done at scales comparable to image-level pretraining. Here, we scale up detection data with self-training, which uses an existing detector to generate pseudo-box annotations on image-text pairs. Major challenges in scaling self-training are the choice of label space, pseudo-annotation filtering, and training efficie"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.09683","kind":"arxiv","version":3},"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/2306.09683/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":"2306.09683","created_at":"2026-07-05T08:21:39.407460+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.09683v3","created_at":"2026-07-05T08:21:39.407460+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.09683","created_at":"2026-07-05T08:21:39.407460+00:00"},{"alias_kind":"pith_short_12","alias_value":"TQNX5JZL74Q5","created_at":"2026-07-05T08:21:39.407460+00:00"},{"alias_kind":"pith_short_16","alias_value":"TQNX5JZL74Q5FQGC","created_at":"2026-07-05T08:21:39.407460+00:00"},{"alias_kind":"pith_short_8","alias_value":"TQNX5JZL","created_at":"2026-07-05T08:21:39.407460+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.12985","citing_title":"Objects Before Words: Object-First Inductive Biases for Grounding Language in Child-View Video","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2605.19410","citing_title":"Vision Harnessing Agent for Open Ad-hoc Segmentation","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2511.16719","citing_title":"SAM 3: Segment Anything with Concepts","ref_index":93,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TQNX5JZL74Q5FQGCST6WD4H32E","json":"https://pith.science/pith/TQNX5JZL74Q5FQGCST6WD4H32E.json","graph_json":"https://pith.science/api/pith-number/TQNX5JZL74Q5FQGCST6WD4H32E/graph.json","events_json":"https://pith.science/api/pith-number/TQNX5JZL74Q5FQGCST6WD4H32E/events.json","paper":"https://pith.science/paper/TQNX5JZL"},"agent_actions":{"view_html":"https://pith.science/pith/TQNX5JZL74Q5FQGCST6WD4H32E","download_json":"https://pith.science/pith/TQNX5JZL74Q5FQGCST6WD4H32E.json","view_paper":"https://pith.science/paper/TQNX5JZL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.09683&json=true","fetch_graph":"https://pith.science/api/pith-number/TQNX5JZL74Q5FQGCST6WD4H32E/graph.json","fetch_events":"https://pith.science/api/pith-number/TQNX5JZL74Q5FQGCST6WD4H32E/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TQNX5JZL74Q5FQGCST6WD4H32E/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TQNX5JZL74Q5FQGCST6WD4H32E/action/storage_attestation","attest_author":"https://pith.science/pith/TQNX5JZL74Q5FQGCST6WD4H32E/action/author_attestation","sign_citation":"https://pith.science/pith/TQNX5JZL74Q5FQGCST6WD4H32E/action/citation_signature","submit_replication":"https://pith.science/pith/TQNX5JZL74Q5FQGCST6WD4H32E/action/replication_record"}},"created_at":"2026-07-05T08:21:39.407460+00:00","updated_at":"2026-07-05T08:21:39.407460+00:00"}