{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:VWTKXCHZ4KGKMFKFOHZXK5ALHJ","short_pith_number":"pith:VWTKXCHZ","schema_version":"1.0","canonical_sha256":"ada6ab88f9e28ca6154571f375740b3a568d7c10f58360d67212e4439bdb712a","source":{"kind":"arxiv","id":"2203.15371","version":4},"attestation_state":"computed","paper":{"title":"mc-BEiT: Multi-choice Discretization for Image BERT Pre-training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Kun Yi, Ling-Yu Duan, Xiaotong Li, Ying Shan, Yixiao Ge, Zixuan Hu","submitted_at":"2022-03-29T09:08:18Z","abstract_excerpt":"Image BERT pre-training with masked image modeling (MIM) becomes a popular practice to cope with self-supervised representation learning. A seminal work, BEiT, casts MIM as a classification task with a visual vocabulary, tokenizing the continuous visual signals into discrete vision tokens using a pre-learned dVAE. Despite a feasible solution, the improper discretization hinders further improvements of image pre-training. Since image discretization has no ground-truth answers, we believe that the masked patch should not be assigned with a unique token id even if a better tokenizer can be obtain"},"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":"2203.15371","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-03-29T09:08:18Z","cross_cats_sorted":[],"title_canon_sha256":"54ca797665b34cbd9fe01afba048c56869c584f381004cb9d359022f1bfbc6f5","abstract_canon_sha256":"19ac12df18bdc95eec31f7d4866a617961daed83030260280fd37a6f9b748fda"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:44:20.690625Z","signature_b64":"XKjdyyz0EZBQSsJsPxtaaaJf8XbNAvRxpUCLNSH/pgtL/fCNpn1tDdjpj7V4Le+JIV4H0F+9ua2R/ghMHzQVAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ada6ab88f9e28ca6154571f375740b3a568d7c10f58360d67212e4439bdb712a","last_reissued_at":"2026-07-05T04:44:20.690119Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:44:20.690119Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"mc-BEiT: Multi-choice Discretization for Image BERT Pre-training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Kun Yi, Ling-Yu Duan, Xiaotong Li, Ying Shan, Yixiao Ge, Zixuan Hu","submitted_at":"2022-03-29T09:08:18Z","abstract_excerpt":"Image BERT pre-training with masked image modeling (MIM) becomes a popular practice to cope with self-supervised representation learning. A seminal work, BEiT, casts MIM as a classification task with a visual vocabulary, tokenizing the continuous visual signals into discrete vision tokens using a pre-learned dVAE. Despite a feasible solution, the improper discretization hinders further improvements of image pre-training. Since image discretization has no ground-truth answers, we believe that the masked patch should not be assigned with a unique token id even if a better tokenizer can be obtain"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.15371","kind":"arxiv","version":4},"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/2203.15371/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":"2203.15371","created_at":"2026-07-05T04:44:20.690188+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.15371v4","created_at":"2026-07-05T04:44:20.690188+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.15371","created_at":"2026-07-05T04:44:20.690188+00:00"},{"alias_kind":"pith_short_12","alias_value":"VWTKXCHZ4KGK","created_at":"2026-07-05T04:44:20.690188+00:00"},{"alias_kind":"pith_short_16","alias_value":"VWTKXCHZ4KGKMFKF","created_at":"2026-07-05T04:44:20.690188+00:00"},{"alias_kind":"pith_short_8","alias_value":"VWTKXCHZ","created_at":"2026-07-05T04:44:20.690188+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/VWTKXCHZ4KGKMFKFOHZXK5ALHJ","json":"https://pith.science/pith/VWTKXCHZ4KGKMFKFOHZXK5ALHJ.json","graph_json":"https://pith.science/api/pith-number/VWTKXCHZ4KGKMFKFOHZXK5ALHJ/graph.json","events_json":"https://pith.science/api/pith-number/VWTKXCHZ4KGKMFKFOHZXK5ALHJ/events.json","paper":"https://pith.science/paper/VWTKXCHZ"},"agent_actions":{"view_html":"https://pith.science/pith/VWTKXCHZ4KGKMFKFOHZXK5ALHJ","download_json":"https://pith.science/pith/VWTKXCHZ4KGKMFKFOHZXK5ALHJ.json","view_paper":"https://pith.science/paper/VWTKXCHZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.15371&json=true","fetch_graph":"https://pith.science/api/pith-number/VWTKXCHZ4KGKMFKFOHZXK5ALHJ/graph.json","fetch_events":"https://pith.science/api/pith-number/VWTKXCHZ4KGKMFKFOHZXK5ALHJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VWTKXCHZ4KGKMFKFOHZXK5ALHJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VWTKXCHZ4KGKMFKFOHZXK5ALHJ/action/storage_attestation","attest_author":"https://pith.science/pith/VWTKXCHZ4KGKMFKFOHZXK5ALHJ/action/author_attestation","sign_citation":"https://pith.science/pith/VWTKXCHZ4KGKMFKFOHZXK5ALHJ/action/citation_signature","submit_replication":"https://pith.science/pith/VWTKXCHZ4KGKMFKFOHZXK5ALHJ/action/replication_record"}},"created_at":"2026-07-05T04:44:20.690188+00:00","updated_at":"2026-07-05T04:44:20.690188+00:00"}