{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:VIVATXEGJLSF33WC4TJRMISVJN","short_pith_number":"pith:VIVATXEG","schema_version":"1.0","canonical_sha256":"aa2a09dc864ae45deec2e4d31622554b6f2e47b948424d9cfad0d9b5ca49b25b","source":{"kind":"arxiv","id":"2107.00249","version":2},"attestation_state":"computed","paper":{"title":"OPT: Omni-Perception Pre-Trainer for Cross-Modal Understanding and Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Fei Liu, Hanqing Lu, Jiajun Zhang, Jing Liu, Jinqiao Wang, Longteng Guo, Mingzhen Sun, Shiyu Zhou, Weining Wang, Xinxin Zhu, Zijia Zhao","submitted_at":"2021-07-01T06:59:44Z","abstract_excerpt":"In this paper, we propose an Omni-perception Pre-Trainer (OPT) for cross-modal understanding and generation, by jointly modeling visual, text and audio resources. OPT is constructed in an encoder-decoder framework, including three single-modal encoders to generate token-based embeddings for each modality, a cross-modal encoder to encode the correlations among the three modalities, and two cross-modal decoders to generate text and image respectively. For the OPT's pre-training, we design a multi-task pretext learning scheme to model multi-modal resources from three different data granularities,"},"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":"2107.00249","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2021-07-01T06:59:44Z","cross_cats_sorted":[],"title_canon_sha256":"ec2d37a3501a6b6ccd19f7f25c6b57678ca020d07bb7f7ba570b2921c2f8458f","abstract_canon_sha256":"853ab4acb93767622b78f82f2c24cc9caeecaf82a06ef17aedf2f530526278d8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:55:17.258302Z","signature_b64":"QdKEycsLMbKO819m3oDNK+OZ4d5xOR/lAeCk9KqHf0rboc5MLDLpsnouI46V7z+dDk6UZS6m2hImq3KWs8EkAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"aa2a09dc864ae45deec2e4d31622554b6f2e47b948424d9cfad0d9b5ca49b25b","last_reissued_at":"2026-07-05T02:55:17.257833Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:55:17.257833Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"OPT: Omni-Perception Pre-Trainer for Cross-Modal Understanding and Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Fei Liu, Hanqing Lu, Jiajun Zhang, Jing Liu, Jinqiao Wang, Longteng Guo, Mingzhen Sun, Shiyu Zhou, Weining Wang, Xinxin Zhu, Zijia Zhao","submitted_at":"2021-07-01T06:59:44Z","abstract_excerpt":"In this paper, we propose an Omni-perception Pre-Trainer (OPT) for cross-modal understanding and generation, by jointly modeling visual, text and audio resources. OPT is constructed in an encoder-decoder framework, including three single-modal encoders to generate token-based embeddings for each modality, a cross-modal encoder to encode the correlations among the three modalities, and two cross-modal decoders to generate text and image respectively. For the OPT's pre-training, we design a multi-task pretext learning scheme to model multi-modal resources from three different data granularities,"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.00249","kind":"arxiv","version":2},"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/2107.00249/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":"2107.00249","created_at":"2026-07-05T02:55:17.257892+00:00"},{"alias_kind":"arxiv_version","alias_value":"2107.00249v2","created_at":"2026-07-05T02:55:17.257892+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.00249","created_at":"2026-07-05T02:55:17.257892+00:00"},{"alias_kind":"pith_short_12","alias_value":"VIVATXEGJLSF","created_at":"2026-07-05T02:55:17.257892+00:00"},{"alias_kind":"pith_short_16","alias_value":"VIVATXEGJLSF33WC","created_at":"2026-07-05T02:55:17.257892+00:00"},{"alias_kind":"pith_short_8","alias_value":"VIVATXEG","created_at":"2026-07-05T02:55:17.257892+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.13364","citing_title":"OmniVec2 -- A Novel Transformer based Network for Large Scale Multimodal and Multitask Learning","ref_index":49,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VIVATXEGJLSF33WC4TJRMISVJN","json":"https://pith.science/pith/VIVATXEGJLSF33WC4TJRMISVJN.json","graph_json":"https://pith.science/api/pith-number/VIVATXEGJLSF33WC4TJRMISVJN/graph.json","events_json":"https://pith.science/api/pith-number/VIVATXEGJLSF33WC4TJRMISVJN/events.json","paper":"https://pith.science/paper/VIVATXEG"},"agent_actions":{"view_html":"https://pith.science/pith/VIVATXEGJLSF33WC4TJRMISVJN","download_json":"https://pith.science/pith/VIVATXEGJLSF33WC4TJRMISVJN.json","view_paper":"https://pith.science/paper/VIVATXEG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2107.00249&json=true","fetch_graph":"https://pith.science/api/pith-number/VIVATXEGJLSF33WC4TJRMISVJN/graph.json","fetch_events":"https://pith.science/api/pith-number/VIVATXEGJLSF33WC4TJRMISVJN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VIVATXEGJLSF33WC4TJRMISVJN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VIVATXEGJLSF33WC4TJRMISVJN/action/storage_attestation","attest_author":"https://pith.science/pith/VIVATXEGJLSF33WC4TJRMISVJN/action/author_attestation","sign_citation":"https://pith.science/pith/VIVATXEGJLSF33WC4TJRMISVJN/action/citation_signature","submit_replication":"https://pith.science/pith/VIVATXEGJLSF33WC4TJRMISVJN/action/replication_record"}},"created_at":"2026-07-05T02:55:17.257892+00:00","updated_at":"2026-07-05T02:55:17.257892+00:00"}