{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:KJ3VHAI33FAQER3ZACFEXLJ5SP","short_pith_number":"pith:KJ3VHAI3","schema_version":"1.0","canonical_sha256":"527753811bd941024779008a4bad3d93fe8e1062aab81f0117fe54061e267b2f","source":{"kind":"arxiv","id":"2303.02995","version":1},"attestation_state":"computed","paper":{"title":"HiCLIP: Contrastive Language-Image Pretraining with Hierarchy-aware Attention","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Jianbo Yuan, Shijie Geng, Yongfeng Zhang, Yu Tian, Yuxiao Chen","submitted_at":"2023-03-06T09:44:01Z","abstract_excerpt":"The success of large-scale contrastive vision-language pretraining (CLIP) has benefited both visual recognition and multimodal content understanding. The concise design brings CLIP the advantage in inference efficiency against other vision-language models with heavier cross-attention fusion layers, making it a popular choice for a wide spectrum of downstream tasks. However, CLIP does not explicitly capture the hierarchical nature of high-level and fine-grained semantics conveyed in images and texts, which is arguably critical to vision-language understanding and reasoning. To this end, we equi"},"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":"2303.02995","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-03-06T09:44:01Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"075d4955127c73c9948b8fd88390ae8f6b1b12e4d83df265e4a3bbae856ddabb","abstract_canon_sha256":"ad250daebb343855e94fb42ab5db55409b46b73ab33bf8ed10a2876b4858929c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:48:18.056625Z","signature_b64":"snhz0lF1JNu5mVvHfAtI8j/6UpSDOSje2XD9YLL9GLT4Ym7Li829FRdfslkCAHUPhnhMbS5exC4HZ8ZNECgoBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"527753811bd941024779008a4bad3d93fe8e1062aab81f0117fe54061e267b2f","last_reissued_at":"2026-07-05T05:48:18.056182Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:48:18.056182Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"HiCLIP: Contrastive Language-Image Pretraining with Hierarchy-aware Attention","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Jianbo Yuan, Shijie Geng, Yongfeng Zhang, Yu Tian, Yuxiao Chen","submitted_at":"2023-03-06T09:44:01Z","abstract_excerpt":"The success of large-scale contrastive vision-language pretraining (CLIP) has benefited both visual recognition and multimodal content understanding. The concise design brings CLIP the advantage in inference efficiency against other vision-language models with heavier cross-attention fusion layers, making it a popular choice for a wide spectrum of downstream tasks. However, CLIP does not explicitly capture the hierarchical nature of high-level and fine-grained semantics conveyed in images and texts, which is arguably critical to vision-language understanding and reasoning. To this end, we equi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.02995","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/2303.02995/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":"2303.02995","created_at":"2026-07-05T05:48:18.056242+00:00"},{"alias_kind":"arxiv_version","alias_value":"2303.02995v1","created_at":"2026-07-05T05:48:18.056242+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.02995","created_at":"2026-07-05T05:48:18.056242+00:00"},{"alias_kind":"pith_short_12","alias_value":"KJ3VHAI33FAQ","created_at":"2026-07-05T05:48:18.056242+00:00"},{"alias_kind":"pith_short_16","alias_value":"KJ3VHAI33FAQER3Z","created_at":"2026-07-05T05:48:18.056242+00:00"},{"alias_kind":"pith_short_8","alias_value":"KJ3VHAI3","created_at":"2026-07-05T05:48:18.056242+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.05268","citing_title":"Is the Geometry Doing the Work? An Operating-Point Audit of Hierarchy in Hyperbolic Vision-Language Models","ref_index":50,"is_internal_anchor":true},{"citing_arxiv_id":"2605.12671","citing_title":"All Circuits Lead to Rome: Rethinking Functional Anisotropy in Circuit and Sheaf Discovery for LLMs","ref_index":298,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KJ3VHAI33FAQER3ZACFEXLJ5SP","json":"https://pith.science/pith/KJ3VHAI33FAQER3ZACFEXLJ5SP.json","graph_json":"https://pith.science/api/pith-number/KJ3VHAI33FAQER3ZACFEXLJ5SP/graph.json","events_json":"https://pith.science/api/pith-number/KJ3VHAI33FAQER3ZACFEXLJ5SP/events.json","paper":"https://pith.science/paper/KJ3VHAI3"},"agent_actions":{"view_html":"https://pith.science/pith/KJ3VHAI33FAQER3ZACFEXLJ5SP","download_json":"https://pith.science/pith/KJ3VHAI33FAQER3ZACFEXLJ5SP.json","view_paper":"https://pith.science/paper/KJ3VHAI3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2303.02995&json=true","fetch_graph":"https://pith.science/api/pith-number/KJ3VHAI33FAQER3ZACFEXLJ5SP/graph.json","fetch_events":"https://pith.science/api/pith-number/KJ3VHAI33FAQER3ZACFEXLJ5SP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KJ3VHAI33FAQER3ZACFEXLJ5SP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KJ3VHAI33FAQER3ZACFEXLJ5SP/action/storage_attestation","attest_author":"https://pith.science/pith/KJ3VHAI33FAQER3ZACFEXLJ5SP/action/author_attestation","sign_citation":"https://pith.science/pith/KJ3VHAI33FAQER3ZACFEXLJ5SP/action/citation_signature","submit_replication":"https://pith.science/pith/KJ3VHAI33FAQER3ZACFEXLJ5SP/action/replication_record"}},"created_at":"2026-07-05T05:48:18.056242+00:00","updated_at":"2026-07-05T05:48:18.056242+00:00"}