{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:F4NB5GMEFTNABLP6FVYIW72MAH","short_pith_number":"pith:F4NB5GME","schema_version":"1.0","canonical_sha256":"2f1a1e99842cda00adfe2d708b7f4c01ebe7653b3ecf371819de2886f5169256","source":{"kind":"arxiv","id":"2203.02378","version":3},"attestation_state":"computed","paper":{"title":"DiT: Self-supervised Pre-training for Document Image Transformer","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Cha Zhang, Furu Wei, Junlong Li, Lei Cui, Tengchao Lv, Yiheng Xu","submitted_at":"2022-03-04T15:34:46Z","abstract_excerpt":"Image Transformer has recently achieved significant progress for natural image understanding, either using supervised (ViT, DeiT, etc.) or self-supervised (BEiT, MAE, etc.) pre-training techniques. In this paper, we propose \\textbf{DiT}, a self-supervised pre-trained \\textbf{D}ocument \\textbf{I}mage \\textbf{T}ransformer model using large-scale unlabeled text images for Document AI tasks, which is essential since no supervised counterparts ever exist due to the lack of human-labeled document images. We leverage DiT as the backbone network in a variety of vision-based Document AI tasks, includin"},"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":"2203.02378","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2022-03-04T15:34:46Z","cross_cats_sorted":[],"title_canon_sha256":"4cdb7486bc9500ba73d9c62673414eceb61245d79661d10348d4c9b9709c8d7d","abstract_canon_sha256":"8c4b87e095ae4461d7d48e62ea35e14177371096616d1506a47514cc1bed90e9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:41:11.505258Z","signature_b64":"hVs44Qwbz71hPqF4z/gkiU5yZO3EGsPlYLKHFRiYFUr7uz+LzPHzA3+LEEVPrDYtlP9xQc+bEflxi7nkJVvxDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2f1a1e99842cda00adfe2d708b7f4c01ebe7653b3ecf371819de2886f5169256","last_reissued_at":"2026-07-05T04:41:11.504802Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:41:11.504802Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DiT: Self-supervised Pre-training for Document Image Transformer","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Cha Zhang, Furu Wei, Junlong Li, Lei Cui, Tengchao Lv, Yiheng Xu","submitted_at":"2022-03-04T15:34:46Z","abstract_excerpt":"Image Transformer has recently achieved significant progress for natural image understanding, either using supervised (ViT, DeiT, etc.) or self-supervised (BEiT, MAE, etc.) pre-training techniques. In this paper, we propose \\textbf{DiT}, a self-supervised pre-trained \\textbf{D}ocument \\textbf{I}mage \\textbf{T}ransformer model using large-scale unlabeled text images for Document AI tasks, which is essential since no supervised counterparts ever exist due to the lack of human-labeled document images. We leverage DiT as the backbone network in a variety of vision-based Document AI tasks, includin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.02378","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/2203.02378/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":"2203.02378","created_at":"2026-07-05T04:41:11.504857+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.02378v3","created_at":"2026-07-05T04:41:11.504857+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.02378","created_at":"2026-07-05T04:41:11.504857+00:00"},{"alias_kind":"pith_short_12","alias_value":"F4NB5GMEFTNA","created_at":"2026-07-05T04:41:11.504857+00:00"},{"alias_kind":"pith_short_16","alias_value":"F4NB5GMEFTNABLP6","created_at":"2026-07-05T04:41:11.504857+00:00"},{"alias_kind":"pith_short_8","alias_value":"F4NB5GME","created_at":"2026-07-05T04:41:11.504857+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2310.05737","citing_title":"Language Model Beats Diffusion -- Tokenizer is Key to Visual Generation","ref_index":195,"is_internal_anchor":false},{"citing_arxiv_id":"2604.11042","citing_title":"Improving Layout Representation Learning Across Inconsistently Annotated Datasets via Agentic Harmonization","ref_index":40,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/F4NB5GMEFTNABLP6FVYIW72MAH","json":"https://pith.science/pith/F4NB5GMEFTNABLP6FVYIW72MAH.json","graph_json":"https://pith.science/api/pith-number/F4NB5GMEFTNABLP6FVYIW72MAH/graph.json","events_json":"https://pith.science/api/pith-number/F4NB5GMEFTNABLP6FVYIW72MAH/events.json","paper":"https://pith.science/paper/F4NB5GME"},"agent_actions":{"view_html":"https://pith.science/pith/F4NB5GMEFTNABLP6FVYIW72MAH","download_json":"https://pith.science/pith/F4NB5GMEFTNABLP6FVYIW72MAH.json","view_paper":"https://pith.science/paper/F4NB5GME","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.02378&json=true","fetch_graph":"https://pith.science/api/pith-number/F4NB5GMEFTNABLP6FVYIW72MAH/graph.json","fetch_events":"https://pith.science/api/pith-number/F4NB5GMEFTNABLP6FVYIW72MAH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/F4NB5GMEFTNABLP6FVYIW72MAH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/F4NB5GMEFTNABLP6FVYIW72MAH/action/storage_attestation","attest_author":"https://pith.science/pith/F4NB5GMEFTNABLP6FVYIW72MAH/action/author_attestation","sign_citation":"https://pith.science/pith/F4NB5GMEFTNABLP6FVYIW72MAH/action/citation_signature","submit_replication":"https://pith.science/pith/F4NB5GMEFTNABLP6FVYIW72MAH/action/replication_record"}},"created_at":"2026-07-05T04:41:11.504857+00:00","updated_at":"2026-07-05T04:41:11.504857+00:00"}