{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:HAAHC77EXKZRA2AB2MXI5XO5XQ","short_pith_number":"pith:HAAHC77E","schema_version":"1.0","canonical_sha256":"3800717fe4bab3106801d32e8eddddbc181f09a17723429032003a6bf4039900","source":{"kind":"arxiv","id":"2203.00242","version":1},"attestation_state":"computed","paper":{"title":"Unsupervised Vision-and-Language Pre-training via Retrieval-based Multi-Granular Alignment","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.MM"],"primary_cat":"cs.CV","authors_text":"Amanpreet Singh, Licheng Yu, Mengjiao Wang, Mingyang Zhou, Ning Zhang, Zhou Yu","submitted_at":"2022-03-01T05:34:01Z","abstract_excerpt":"Vision-and-Language (V+L) pre-training models have achieved tremendous success in recent years on various multi-modal benchmarks. However, the majority of existing models require pre-training on a large set of parallel image-text data, which is costly to collect, compared to image-only or text-only data. In this paper, we explore unsupervised Vision-and-Language pre-training (UVLP) to learn the cross-modal representation from non-parallel image and text datasets. We found two key factors that lead to good unsupervised V+L pre-training without parallel data: (i) joint image-and-text input (ii) "},"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":"2203.00242","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-03-01T05:34:01Z","cross_cats_sorted":["cs.AI","cs.CL","cs.MM"],"title_canon_sha256":"70ed288465bd36d1b35b423ba4c6da079fe2d65c82927d8acbf30df5623d7ead","abstract_canon_sha256":"52fd1634491ec0b9e38d1972a048ba62572cae90f572e65f00c89ca054d14750"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:01:15.452155Z","signature_b64":"qAn1UC4RlPGm68Cjv7DgSb6SM+N5wI4nLDsGFy23iOxnymVmzYWoOGAK95sK0/mQi85E7/5wNvTma44eCRqmDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3800717fe4bab3106801d32e8eddddbc181f09a17723429032003a6bf4039900","last_reissued_at":"2026-07-05T04:01:15.451738Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:01:15.451738Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Unsupervised Vision-and-Language Pre-training via Retrieval-based Multi-Granular Alignment","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.MM"],"primary_cat":"cs.CV","authors_text":"Amanpreet Singh, Licheng Yu, Mengjiao Wang, Mingyang Zhou, Ning Zhang, Zhou Yu","submitted_at":"2022-03-01T05:34:01Z","abstract_excerpt":"Vision-and-Language (V+L) pre-training models have achieved tremendous success in recent years on various multi-modal benchmarks. However, the majority of existing models require pre-training on a large set of parallel image-text data, which is costly to collect, compared to image-only or text-only data. In this paper, we explore unsupervised Vision-and-Language pre-training (UVLP) to learn the cross-modal representation from non-parallel image and text datasets. We found two key factors that lead to good unsupervised V+L pre-training without parallel data: (i) joint image-and-text input (ii) "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.00242","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/2203.00242/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":"2203.00242","created_at":"2026-07-05T04:01:15.451796+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.00242v1","created_at":"2026-07-05T04:01:15.451796+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.00242","created_at":"2026-07-05T04:01:15.451796+00:00"},{"alias_kind":"pith_short_12","alias_value":"HAAHC77EXKZR","created_at":"2026-07-05T04:01:15.451796+00:00"},{"alias_kind":"pith_short_16","alias_value":"HAAHC77EXKZRA2AB","created_at":"2026-07-05T04:01:15.451796+00:00"},{"alias_kind":"pith_short_8","alias_value":"HAAHC77E","created_at":"2026-07-05T04:01:15.451796+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/HAAHC77EXKZRA2AB2MXI5XO5XQ","json":"https://pith.science/pith/HAAHC77EXKZRA2AB2MXI5XO5XQ.json","graph_json":"https://pith.science/api/pith-number/HAAHC77EXKZRA2AB2MXI5XO5XQ/graph.json","events_json":"https://pith.science/api/pith-number/HAAHC77EXKZRA2AB2MXI5XO5XQ/events.json","paper":"https://pith.science/paper/HAAHC77E"},"agent_actions":{"view_html":"https://pith.science/pith/HAAHC77EXKZRA2AB2MXI5XO5XQ","download_json":"https://pith.science/pith/HAAHC77EXKZRA2AB2MXI5XO5XQ.json","view_paper":"https://pith.science/paper/HAAHC77E","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.00242&json=true","fetch_graph":"https://pith.science/api/pith-number/HAAHC77EXKZRA2AB2MXI5XO5XQ/graph.json","fetch_events":"https://pith.science/api/pith-number/HAAHC77EXKZRA2AB2MXI5XO5XQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HAAHC77EXKZRA2AB2MXI5XO5XQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HAAHC77EXKZRA2AB2MXI5XO5XQ/action/storage_attestation","attest_author":"https://pith.science/pith/HAAHC77EXKZRA2AB2MXI5XO5XQ/action/author_attestation","sign_citation":"https://pith.science/pith/HAAHC77EXKZRA2AB2MXI5XO5XQ/action/citation_signature","submit_replication":"https://pith.science/pith/HAAHC77EXKZRA2AB2MXI5XO5XQ/action/replication_record"}},"created_at":"2026-07-05T04:01:15.451796+00:00","updated_at":"2026-07-05T04:01:15.451796+00:00"}