{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:YEXI5ZX5UMDSCHUWMSJG5EE77S","short_pith_number":"pith:YEXI5ZX5","schema_version":"1.0","canonical_sha256":"c12e8ee6fda307211e9664926e909ffcbc27a5e0ef579554abdda9d6ae7347e1","source":{"kind":"arxiv","id":"2103.07829","version":1},"attestation_state":"computed","paper":{"title":"SemVLP: Vision-Language Pre-training by Aligning Semantics at Multiple Levels","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bin Bi, Chenliang Li, Fuli Luo, Haiyang Xu, Ming Yan, Songfang Huang, Wei Wang","submitted_at":"2021-03-14T02:39:14Z","abstract_excerpt":"Vision-language pre-training (VLP) on large-scale image-text pairs has recently witnessed rapid progress for learning cross-modal representations. Existing pre-training methods either directly concatenate image representation and text representation at a feature level as input to a single-stream Transformer, or use a two-stream cross-modal Transformer to align the image-text representation at a high-level semantic space. In real-world image-text data, we observe that it is easy for some of the image-text pairs to align simple semantics on both modalities, while others may be related after high"},"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":"2103.07829","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-03-14T02:39:14Z","cross_cats_sorted":[],"title_canon_sha256":"11057b2769262e792e6d398c1badd63216a2b363de0a7d818da2bac56989e84d","abstract_canon_sha256":"29f4c3fbee01c9d82fdfd1460600622601ccba8bc785a7f091b02c54ea21773f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:22:45.438483Z","signature_b64":"hbE9oGMNJwhZ/p4PgpmXddCN5QrjUEnb7KNb85usATuY5mUxVOKYVjf/AecTGrRANCujPnBDmzlR4Qc5I4l3Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c12e8ee6fda307211e9664926e909ffcbc27a5e0ef579554abdda9d6ae7347e1","last_reissued_at":"2026-07-05T02:22:45.438062Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:22:45.438062Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SemVLP: Vision-Language Pre-training by Aligning Semantics at Multiple Levels","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bin Bi, Chenliang Li, Fuli Luo, Haiyang Xu, Ming Yan, Songfang Huang, Wei Wang","submitted_at":"2021-03-14T02:39:14Z","abstract_excerpt":"Vision-language pre-training (VLP) on large-scale image-text pairs has recently witnessed rapid progress for learning cross-modal representations. Existing pre-training methods either directly concatenate image representation and text representation at a feature level as input to a single-stream Transformer, or use a two-stream cross-modal Transformer to align the image-text representation at a high-level semantic space. In real-world image-text data, we observe that it is easy for some of the image-text pairs to align simple semantics on both modalities, while others may be related after high"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2103.07829","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/2103.07829/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":"2103.07829","created_at":"2026-07-05T02:22:45.438125+00:00"},{"alias_kind":"arxiv_version","alias_value":"2103.07829v1","created_at":"2026-07-05T02:22:45.438125+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2103.07829","created_at":"2026-07-05T02:22:45.438125+00:00"},{"alias_kind":"pith_short_12","alias_value":"YEXI5ZX5UMDS","created_at":"2026-07-05T02:22:45.438125+00:00"},{"alias_kind":"pith_short_16","alias_value":"YEXI5ZX5UMDSCHUW","created_at":"2026-07-05T02:22:45.438125+00:00"},{"alias_kind":"pith_short_8","alias_value":"YEXI5ZX5","created_at":"2026-07-05T02:22:45.438125+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/YEXI5ZX5UMDSCHUWMSJG5EE77S","json":"https://pith.science/pith/YEXI5ZX5UMDSCHUWMSJG5EE77S.json","graph_json":"https://pith.science/api/pith-number/YEXI5ZX5UMDSCHUWMSJG5EE77S/graph.json","events_json":"https://pith.science/api/pith-number/YEXI5ZX5UMDSCHUWMSJG5EE77S/events.json","paper":"https://pith.science/paper/YEXI5ZX5"},"agent_actions":{"view_html":"https://pith.science/pith/YEXI5ZX5UMDSCHUWMSJG5EE77S","download_json":"https://pith.science/pith/YEXI5ZX5UMDSCHUWMSJG5EE77S.json","view_paper":"https://pith.science/paper/YEXI5ZX5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2103.07829&json=true","fetch_graph":"https://pith.science/api/pith-number/YEXI5ZX5UMDSCHUWMSJG5EE77S/graph.json","fetch_events":"https://pith.science/api/pith-number/YEXI5ZX5UMDSCHUWMSJG5EE77S/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YEXI5ZX5UMDSCHUWMSJG5EE77S/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YEXI5ZX5UMDSCHUWMSJG5EE77S/action/storage_attestation","attest_author":"https://pith.science/pith/YEXI5ZX5UMDSCHUWMSJG5EE77S/action/author_attestation","sign_citation":"https://pith.science/pith/YEXI5ZX5UMDSCHUWMSJG5EE77S/action/citation_signature","submit_replication":"https://pith.science/pith/YEXI5ZX5UMDSCHUWMSJG5EE77S/action/replication_record"}},"created_at":"2026-07-05T02:22:45.438125+00:00","updated_at":"2026-07-05T02:22:45.438125+00:00"}