{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:PEF6GYZSXEFOVCQBD24G3XKJFJ","short_pith_number":"pith:PEF6GYZS","schema_version":"1.0","canonical_sha256":"790be36332b90aea8a011eb86ddd492a549cbd25d00bea6265605ce0846240f2","source":{"kind":"arxiv","id":"1909.11059","version":3},"attestation_state":"computed","paper":{"title":"Unified Vision-Language Pre-Training for Image Captioning and VQA","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hamid Palangi, Houdong Hu, Jason J. Corso, Jianfeng Gao, Lei Zhang, Luowei Zhou","submitted_at":"2019-09-24T17:17:26Z","abstract_excerpt":"This paper presents a unified Vision-Language Pre-training (VLP) model. The model is unified in that (1) it can be fine-tuned for either vision-language generation (e.g., image captioning) or understanding (e.g., visual question answering) tasks, and (2) it uses a shared multi-layer transformer network for both encoding and decoding, which differs from many existing methods where the encoder and decoder are implemented using separate models. The unified VLP model is pre-trained on a large amount of image-text pairs using the unsupervised learning objectives of two tasks: bidirectional and sequ"},"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":"1909.11059","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2019-09-24T17:17:26Z","cross_cats_sorted":[],"title_canon_sha256":"40f262b6130f4bd7b652b9b0d04cdfa8b85fff68364a2e0c39c53ea73ece774d","abstract_canon_sha256":"bc86dd252dc8ed241dea637e6d294b77731074e7e266cf6859635f46344754c4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:23:51.326924Z","signature_b64":"fJkFZNVQYUvM08afr9wEBu/XAEtKAWaN9vZHlOKtShnfqPb4gXhKQ0jjU9rD4M/4ygkeV0HszRY7H6SnpZbLDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"790be36332b90aea8a011eb86ddd492a549cbd25d00bea6265605ce0846240f2","last_reissued_at":"2026-07-05T00:23:51.326420Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:23:51.326420Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Unified Vision-Language Pre-Training for Image Captioning and VQA","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hamid Palangi, Houdong Hu, Jason J. Corso, Jianfeng Gao, Lei Zhang, Luowei Zhou","submitted_at":"2019-09-24T17:17:26Z","abstract_excerpt":"This paper presents a unified Vision-Language Pre-training (VLP) model. The model is unified in that (1) it can be fine-tuned for either vision-language generation (e.g., image captioning) or understanding (e.g., visual question answering) tasks, and (2) it uses a shared multi-layer transformer network for both encoding and decoding, which differs from many existing methods where the encoder and decoder are implemented using separate models. The unified VLP model is pre-trained on a large amount of image-text pairs using the unsupervised learning objectives of two tasks: bidirectional and sequ"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1909.11059","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/1909.11059/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":"1909.11059","created_at":"2026-07-05T00:23:51.326484+00:00"},{"alias_kind":"arxiv_version","alias_value":"1909.11059v3","created_at":"2026-07-05T00:23:51.326484+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1909.11059","created_at":"2026-07-05T00:23:51.326484+00:00"},{"alias_kind":"pith_short_12","alias_value":"PEF6GYZSXEFO","created_at":"2026-07-05T00:23:51.326484+00:00"},{"alias_kind":"pith_short_16","alias_value":"PEF6GYZSXEFOVCQB","created_at":"2026-07-05T00:23:51.326484+00:00"},{"alias_kind":"pith_short_8","alias_value":"PEF6GYZS","created_at":"2026-07-05T00:23:51.326484+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2308.01390","citing_title":"OpenFlamingo: An Open-Source Framework for Training Large Autoregressive Vision-Language Models","ref_index":44,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PEF6GYZSXEFOVCQBD24G3XKJFJ","json":"https://pith.science/pith/PEF6GYZSXEFOVCQBD24G3XKJFJ.json","graph_json":"https://pith.science/api/pith-number/PEF6GYZSXEFOVCQBD24G3XKJFJ/graph.json","events_json":"https://pith.science/api/pith-number/PEF6GYZSXEFOVCQBD24G3XKJFJ/events.json","paper":"https://pith.science/paper/PEF6GYZS"},"agent_actions":{"view_html":"https://pith.science/pith/PEF6GYZSXEFOVCQBD24G3XKJFJ","download_json":"https://pith.science/pith/PEF6GYZSXEFOVCQBD24G3XKJFJ.json","view_paper":"https://pith.science/paper/PEF6GYZS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1909.11059&json=true","fetch_graph":"https://pith.science/api/pith-number/PEF6GYZSXEFOVCQBD24G3XKJFJ/graph.json","fetch_events":"https://pith.science/api/pith-number/PEF6GYZSXEFOVCQBD24G3XKJFJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PEF6GYZSXEFOVCQBD24G3XKJFJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PEF6GYZSXEFOVCQBD24G3XKJFJ/action/storage_attestation","attest_author":"https://pith.science/pith/PEF6GYZSXEFOVCQBD24G3XKJFJ/action/author_attestation","sign_citation":"https://pith.science/pith/PEF6GYZSXEFOVCQBD24G3XKJFJ/action/citation_signature","submit_replication":"https://pith.science/pith/PEF6GYZSXEFOVCQBD24G3XKJFJ/action/replication_record"}},"created_at":"2026-07-05T00:23:51.326484+00:00","updated_at":"2026-07-05T00:23:51.326484+00:00"}