{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:TYZ7PF4JLUYEBFLF7YRMGPLU5C","short_pith_number":"pith:TYZ7PF4J","schema_version":"1.0","canonical_sha256":"9e33f797895d30409565fe22c33d74e8a50d0a05598b044fc7e071ca20694c0a","source":{"kind":"arxiv","id":"2202.10401","version":4},"attestation_state":"computed","paper":{"title":"Vision-Language Pre-Training with Triple Contrastive Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Belinda Zeng, Jiali Duan, Jinyu Yang, Junzhou Huang, Liqun Chen, Sampath Chanda, Son Tran, Trishul Chilimbi, Yi Xu","submitted_at":"2022-02-21T17:54:57Z","abstract_excerpt":"Vision-language representation learning largely benefits from image-text alignment through contrastive losses (e.g., InfoNCE loss). The success of this alignment strategy is attributed to its capability in maximizing the mutual information (MI) between an image and its matched text. However, simply performing cross-modal alignment (CMA) ignores data potential within each modality, which may result in degraded representations. For instance, although CMA-based models are able to map image-text pairs close together in the embedding space, they fail to ensure that similar inputs from the same moda"},"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":"2202.10401","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-02-21T17:54:57Z","cross_cats_sorted":[],"title_canon_sha256":"6ac83b3a7760bd33231ed454ecd141a63f846a72391a1da1a6a18efc35f20fce","abstract_canon_sha256":"edeac47acc9a16371834f5c536463d44fead261b108237a2842e0b018faba92d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:09:01.775422Z","signature_b64":"S+Y4xY0qLxxdORaGOLSCmfScFgcxBKJHDTs57qIJzNNdZn50K0/BiM77bNYmeDAS+EwD/D3N1o+HCSqwp0faDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9e33f797895d30409565fe22c33d74e8a50d0a05598b044fc7e071ca20694c0a","last_reissued_at":"2026-07-05T04:09:01.775025Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:09:01.775025Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Vision-Language Pre-Training with Triple Contrastive Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Belinda Zeng, Jiali Duan, Jinyu Yang, Junzhou Huang, Liqun Chen, Sampath Chanda, Son Tran, Trishul Chilimbi, Yi Xu","submitted_at":"2022-02-21T17:54:57Z","abstract_excerpt":"Vision-language representation learning largely benefits from image-text alignment through contrastive losses (e.g., InfoNCE loss). The success of this alignment strategy is attributed to its capability in maximizing the mutual information (MI) between an image and its matched text. However, simply performing cross-modal alignment (CMA) ignores data potential within each modality, which may result in degraded representations. For instance, although CMA-based models are able to map image-text pairs close together in the embedding space, they fail to ensure that similar inputs from the same moda"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2202.10401","kind":"arxiv","version":4},"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/2202.10401/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":"2202.10401","created_at":"2026-07-05T04:09:01.775082+00:00"},{"alias_kind":"arxiv_version","alias_value":"2202.10401v4","created_at":"2026-07-05T04:09:01.775082+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2202.10401","created_at":"2026-07-05T04:09:01.775082+00:00"},{"alias_kind":"pith_short_12","alias_value":"TYZ7PF4JLUYE","created_at":"2026-07-05T04:09:01.775082+00:00"},{"alias_kind":"pith_short_16","alias_value":"TYZ7PF4JLUYEBFLF","created_at":"2026-07-05T04:09:01.775082+00:00"},{"alias_kind":"pith_short_8","alias_value":"TYZ7PF4J","created_at":"2026-07-05T04:09:01.775082+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TYZ7PF4JLUYEBFLF7YRMGPLU5C","json":"https://pith.science/pith/TYZ7PF4JLUYEBFLF7YRMGPLU5C.json","graph_json":"https://pith.science/api/pith-number/TYZ7PF4JLUYEBFLF7YRMGPLU5C/graph.json","events_json":"https://pith.science/api/pith-number/TYZ7PF4JLUYEBFLF7YRMGPLU5C/events.json","paper":"https://pith.science/paper/TYZ7PF4J"},"agent_actions":{"view_html":"https://pith.science/pith/TYZ7PF4JLUYEBFLF7YRMGPLU5C","download_json":"https://pith.science/pith/TYZ7PF4JLUYEBFLF7YRMGPLU5C.json","view_paper":"https://pith.science/paper/TYZ7PF4J","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2202.10401&json=true","fetch_graph":"https://pith.science/api/pith-number/TYZ7PF4JLUYEBFLF7YRMGPLU5C/graph.json","fetch_events":"https://pith.science/api/pith-number/TYZ7PF4JLUYEBFLF7YRMGPLU5C/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TYZ7PF4JLUYEBFLF7YRMGPLU5C/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TYZ7PF4JLUYEBFLF7YRMGPLU5C/action/storage_attestation","attest_author":"https://pith.science/pith/TYZ7PF4JLUYEBFLF7YRMGPLU5C/action/author_attestation","sign_citation":"https://pith.science/pith/TYZ7PF4JLUYEBFLF7YRMGPLU5C/action/citation_signature","submit_replication":"https://pith.science/pith/TYZ7PF4JLUYEBFLF7YRMGPLU5C/action/replication_record"}},"created_at":"2026-07-05T04:09:01.775082+00:00","updated_at":"2026-07-05T04:09:01.775082+00:00"}