{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:NDLC6ZRVNVUNHET3UY22E6TO7V","short_pith_number":"pith:NDLC6ZRV","schema_version":"1.0","canonical_sha256":"68d62f66356d68d3927ba635a27a6efd605242100f672650abee9b68f74775ad","source":{"kind":"arxiv","id":"2607.11562","version":1},"attestation_state":"computed","paper":{"title":"MonkeyOCRv2: A Visual-Text Foundation Model for Document AI","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dongliang Luo, Handong Zheng, Jiajun Song, Jiarui Zhang, Qiang Liu, Shuo Zhang, Xiang Bai, Xinhan Wang, Yang Liu, Yuliang Liu, Zhang Li, Zhiyin Ma, Zidun Guo, Ziyang Zhang","submitted_at":"2026-07-13T13:43:39Z","abstract_excerpt":"Mainstream visual encoders are pretrained on natural images and cannot be effectively applied to document images without document-oriented adaptation, as dense text and fine-grained character strokes demand character-level visual perception. We present MonkeyOCRv2, a visual-text pretrained model for document AI. First, we construct MonkeyDoc v2, to our knowledge the largest document-image pretraining corpus, comprising 113 million images spanning 17 languages. Second, we propose a pretraining strategy that jointly learns image-to-text generation and pixel-level document reconstruction: the for"},"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":"2607.11562","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.CV","submitted_at":"2026-07-13T13:43:39Z","cross_cats_sorted":[],"title_canon_sha256":"5ae7015983c6e33aa74968fe790422f26e4e3294f5c43c81d4bf9e504216d2fe","abstract_canon_sha256":"25d05e12407822f2cd839fd8b23d6b5c563765d6b2d2f1bc0aa789ebb99971fc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-14T02:22:12.129691Z","signature_b64":"i/OS0uLRx9GSFin3Y6Yptai47rGvFMnKOt2QvPWfsPhmb63gVYwWAh6k1D5QgKNyNOYkKizm3oVSk2gmae9bAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"68d62f66356d68d3927ba635a27a6efd605242100f672650abee9b68f74775ad","last_reissued_at":"2026-07-14T02:22:12.128880Z","signature_status":"signed_v1","first_computed_at":"2026-07-14T02:22:12.128880Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MonkeyOCRv2: A Visual-Text Foundation Model for Document AI","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dongliang Luo, Handong Zheng, Jiajun Song, Jiarui Zhang, Qiang Liu, Shuo Zhang, Xiang Bai, Xinhan Wang, Yang Liu, Yuliang Liu, Zhang Li, Zhiyin Ma, Zidun Guo, Ziyang Zhang","submitted_at":"2026-07-13T13:43:39Z","abstract_excerpt":"Mainstream visual encoders are pretrained on natural images and cannot be effectively applied to document images without document-oriented adaptation, as dense text and fine-grained character strokes demand character-level visual perception. We present MonkeyOCRv2, a visual-text pretrained model for document AI. First, we construct MonkeyDoc v2, to our knowledge the largest document-image pretraining corpus, comprising 113 million images spanning 17 languages. Second, we propose a pretraining strategy that jointly learns image-to-text generation and pixel-level document reconstruction: the for"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.11562","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/2607.11562/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":"2607.11562","created_at":"2026-07-14T02:22:12.129278+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.11562v1","created_at":"2026-07-14T02:22:12.129278+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.11562","created_at":"2026-07-14T02:22:12.129278+00:00"},{"alias_kind":"pith_short_12","alias_value":"NDLC6ZRVNVUN","created_at":"2026-07-14T02:22:12.129278+00:00"},{"alias_kind":"pith_short_16","alias_value":"NDLC6ZRVNVUNHET3","created_at":"2026-07-14T02:22:12.129278+00:00"},{"alias_kind":"pith_short_8","alias_value":"NDLC6ZRV","created_at":"2026-07-14T02:22:12.129278+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2608.06146","citing_title":"PaDoc: Layout-Grounded Parallel Decoding for Document Parsing","ref_index":11,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NDLC6ZRVNVUNHET3UY22E6TO7V","json":"https://pith.science/pith/NDLC6ZRVNVUNHET3UY22E6TO7V.json","graph_json":"https://pith.science/api/pith-number/NDLC6ZRVNVUNHET3UY22E6TO7V/graph.json","events_json":"https://pith.science/api/pith-number/NDLC6ZRVNVUNHET3UY22E6TO7V/events.json","paper":"https://pith.science/paper/NDLC6ZRV"},"agent_actions":{"view_html":"https://pith.science/pith/NDLC6ZRVNVUNHET3UY22E6TO7V","download_json":"https://pith.science/pith/NDLC6ZRVNVUNHET3UY22E6TO7V.json","view_paper":"https://pith.science/paper/NDLC6ZRV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.11562&json=true","fetch_graph":"https://pith.science/api/pith-number/NDLC6ZRVNVUNHET3UY22E6TO7V/graph.json","fetch_events":"https://pith.science/api/pith-number/NDLC6ZRVNVUNHET3UY22E6TO7V/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NDLC6ZRVNVUNHET3UY22E6TO7V/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NDLC6ZRVNVUNHET3UY22E6TO7V/action/storage_attestation","attest_author":"https://pith.science/pith/NDLC6ZRVNVUNHET3UY22E6TO7V/action/author_attestation","sign_citation":"https://pith.science/pith/NDLC6ZRVNVUNHET3UY22E6TO7V/action/citation_signature","submit_replication":"https://pith.science/pith/NDLC6ZRVNVUNHET3UY22E6TO7V/action/replication_record"}},"created_at":"2026-07-14T02:22:12.129278+00:00","updated_at":"2026-07-14T02:22:12.129278+00:00"}