{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:5VGHJETJJO7IXLYDPBCZC3JMRF","short_pith_number":"pith:5VGHJETJ","schema_version":"1.0","canonical_sha256":"ed4c7492694bbe8baf037845916d2c8944355a6688946d0f1708b262174947e1","source":{"kind":"arxiv","id":"2206.00311","version":3},"attestation_state":"computed","paper":{"title":"MaskOCR: Text Recognition with Masked Encoder-Decoder Pretraining","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chengquan Zhang, Errui Ding, Jingdong Wang, Junyu Han, Kun Yao, Liang Wu, Meina Qiao, Pengyuan Lyu, Shanshan Liu, Yangliu Xu","submitted_at":"2022-06-01T08:27:19Z","abstract_excerpt":"Text images contain both visual and linguistic information. However, existing pre-training techniques for text recognition mainly focus on either visual representation learning or linguistic knowledge learning. In this paper, we propose a novel approach MaskOCR to unify vision and language pre-training in the classical encoder-decoder recognition framework. We adopt the masked image modeling approach to pre-train the feature encoder using a large set of unlabeled real text images, which allows us to learn strong visual representations. In contrast to introducing linguistic knowledge with an ad"},"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":"2206.00311","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-06-01T08:27:19Z","cross_cats_sorted":[],"title_canon_sha256":"6a10620bd80e1aab1b0a8b4cac21631e828037d1f99b88be7fa0823c2f3347c2","abstract_canon_sha256":"17865445fcdc26ef1bf18c2815f17f121b39afcaf06508cdec8ab71117b77691"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:58:53.007145Z","signature_b64":"CE5t307Tod4uKTMOu7cbCMeZa7vgd0pnUhwe3SmnDtb0URwSmOozbewKbqwsNOQc/0GpoUJSp+d/vPhaKyteCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ed4c7492694bbe8baf037845916d2c8944355a6688946d0f1708b262174947e1","last_reissued_at":"2026-07-05T06:58:53.006609Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:58:53.006609Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MaskOCR: Text Recognition with Masked Encoder-Decoder Pretraining","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chengquan Zhang, Errui Ding, Jingdong Wang, Junyu Han, Kun Yao, Liang Wu, Meina Qiao, Pengyuan Lyu, Shanshan Liu, Yangliu Xu","submitted_at":"2022-06-01T08:27:19Z","abstract_excerpt":"Text images contain both visual and linguistic information. However, existing pre-training techniques for text recognition mainly focus on either visual representation learning or linguistic knowledge learning. In this paper, we propose a novel approach MaskOCR to unify vision and language pre-training in the classical encoder-decoder recognition framework. We adopt the masked image modeling approach to pre-train the feature encoder using a large set of unlabeled real text images, which allows us to learn strong visual representations. In contrast to introducing linguistic knowledge with an ad"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2206.00311","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/2206.00311/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":"2206.00311","created_at":"2026-07-05T06:58:53.006668+00:00"},{"alias_kind":"arxiv_version","alias_value":"2206.00311v3","created_at":"2026-07-05T06:58:53.006668+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2206.00311","created_at":"2026-07-05T06:58:53.006668+00:00"},{"alias_kind":"pith_short_12","alias_value":"5VGHJETJJO7I","created_at":"2026-07-05T06:58:53.006668+00:00"},{"alias_kind":"pith_short_16","alias_value":"5VGHJETJJO7IXLYD","created_at":"2026-07-05T06:58:53.006668+00:00"},{"alias_kind":"pith_short_8","alias_value":"5VGHJETJ","created_at":"2026-07-05T06:58:53.006668+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2205.14100","citing_title":"GIT: A Generative Image-to-text Transformer for Vision and Language","ref_index":22,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5VGHJETJJO7IXLYDPBCZC3JMRF","json":"https://pith.science/pith/5VGHJETJJO7IXLYDPBCZC3JMRF.json","graph_json":"https://pith.science/api/pith-number/5VGHJETJJO7IXLYDPBCZC3JMRF/graph.json","events_json":"https://pith.science/api/pith-number/5VGHJETJJO7IXLYDPBCZC3JMRF/events.json","paper":"https://pith.science/paper/5VGHJETJ"},"agent_actions":{"view_html":"https://pith.science/pith/5VGHJETJJO7IXLYDPBCZC3JMRF","download_json":"https://pith.science/pith/5VGHJETJJO7IXLYDPBCZC3JMRF.json","view_paper":"https://pith.science/paper/5VGHJETJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2206.00311&json=true","fetch_graph":"https://pith.science/api/pith-number/5VGHJETJJO7IXLYDPBCZC3JMRF/graph.json","fetch_events":"https://pith.science/api/pith-number/5VGHJETJJO7IXLYDPBCZC3JMRF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5VGHJETJJO7IXLYDPBCZC3JMRF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5VGHJETJJO7IXLYDPBCZC3JMRF/action/storage_attestation","attest_author":"https://pith.science/pith/5VGHJETJJO7IXLYDPBCZC3JMRF/action/author_attestation","sign_citation":"https://pith.science/pith/5VGHJETJJO7IXLYDPBCZC3JMRF/action/citation_signature","submit_replication":"https://pith.science/pith/5VGHJETJJO7IXLYDPBCZC3JMRF/action/replication_record"}},"created_at":"2026-07-05T06:58:53.006668+00:00","updated_at":"2026-07-05T06:58:53.006668+00:00"}