{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:CQNJWC4XV6R5XAIMN47PTWQ4XC","short_pith_number":"pith:CQNJWC4X","schema_version":"1.0","canonical_sha256":"141a9b0b97afa3db810c6f3ef9da1cb8971f90a807654975e7f1a22342908422","source":{"kind":"arxiv","id":"1912.09582","version":1},"attestation_state":"computed","paper":{"title":"BERTje: A Dutch BERT Model","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Andreas van Cranenburgh, Arianna Bisazza, Gertjan van Noord, Malvina Nissim, Tommaso Caselli, Wietse de Vries","submitted_at":"2019-12-19T22:59:26Z","abstract_excerpt":"The transformer-based pre-trained language model BERT has helped to improve state-of-the-art performance on many natural language processing (NLP) tasks. Using the same architecture and parameters, we developed and evaluated a monolingual Dutch BERT model called BERTje. Compared to the multilingual BERT model, which includes Dutch but is only based on Wikipedia text, BERTje is based on a large and diverse dataset of 2.4 billion tokens. BERTje consistently outperforms the equally-sized multilingual BERT model on downstream NLP tasks (part-of-speech tagging, named-entity recognition, semantic ro"},"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":"1912.09582","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2019-12-19T22:59:26Z","cross_cats_sorted":[],"title_canon_sha256":"840b1cf9b4500ba604e4d55c3f25861f6534386f9d061b3ae32d0d685cbbe9e3","abstract_canon_sha256":"10550e08f019656465920cf778d48ea17d1e03862b246da8c08864a9b40334d0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:27:35.753540Z","signature_b64":"OzBS4rlocKyPGXNi7RkkfHh4P+3HuONyswZWmO6XNb0zLWzVkIwq6jPGB/B/6xegOoMZPGAz9tsTzxnnwGMyBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"141a9b0b97afa3db810c6f3ef9da1cb8971f90a807654975e7f1a22342908422","last_reissued_at":"2026-07-05T00:27:35.753050Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:27:35.753050Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"BERTje: A Dutch BERT Model","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Andreas van Cranenburgh, Arianna Bisazza, Gertjan van Noord, Malvina Nissim, Tommaso Caselli, Wietse de Vries","submitted_at":"2019-12-19T22:59:26Z","abstract_excerpt":"The transformer-based pre-trained language model BERT has helped to improve state-of-the-art performance on many natural language processing (NLP) tasks. Using the same architecture and parameters, we developed and evaluated a monolingual Dutch BERT model called BERTje. Compared to the multilingual BERT model, which includes Dutch but is only based on Wikipedia text, BERTje is based on a large and diverse dataset of 2.4 billion tokens. BERTje consistently outperforms the equally-sized multilingual BERT model on downstream NLP tasks (part-of-speech tagging, named-entity recognition, semantic ro"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1912.09582","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/1912.09582/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":"1912.09582","created_at":"2026-07-05T00:27:35.753108+00:00"},{"alias_kind":"arxiv_version","alias_value":"1912.09582v1","created_at":"2026-07-05T00:27:35.753108+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1912.09582","created_at":"2026-07-05T00:27:35.753108+00:00"},{"alias_kind":"pith_short_12","alias_value":"CQNJWC4XV6R5","created_at":"2026-07-05T00:27:35.753108+00:00"},{"alias_kind":"pith_short_16","alias_value":"CQNJWC4XV6R5XAIM","created_at":"2026-07-05T00:27:35.753108+00:00"},{"alias_kind":"pith_short_8","alias_value":"CQNJWC4X","created_at":"2026-07-05T00:27:35.753108+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.02100","citing_title":"PortBERT: Navigating the Depths of Portuguese Language Models","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05902","citing_title":"Evaluating Non-English Developer Support in Machine Learning for Software Engineering","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2604.21421","citing_title":"Differentially Private De-identification of Dutch Clinical Notes: A Comparative Evaluation","ref_index":25,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CQNJWC4XV6R5XAIMN47PTWQ4XC","json":"https://pith.science/pith/CQNJWC4XV6R5XAIMN47PTWQ4XC.json","graph_json":"https://pith.science/api/pith-number/CQNJWC4XV6R5XAIMN47PTWQ4XC/graph.json","events_json":"https://pith.science/api/pith-number/CQNJWC4XV6R5XAIMN47PTWQ4XC/events.json","paper":"https://pith.science/paper/CQNJWC4X"},"agent_actions":{"view_html":"https://pith.science/pith/CQNJWC4XV6R5XAIMN47PTWQ4XC","download_json":"https://pith.science/pith/CQNJWC4XV6R5XAIMN47PTWQ4XC.json","view_paper":"https://pith.science/paper/CQNJWC4X","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1912.09582&json=true","fetch_graph":"https://pith.science/api/pith-number/CQNJWC4XV6R5XAIMN47PTWQ4XC/graph.json","fetch_events":"https://pith.science/api/pith-number/CQNJWC4XV6R5XAIMN47PTWQ4XC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CQNJWC4XV6R5XAIMN47PTWQ4XC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CQNJWC4XV6R5XAIMN47PTWQ4XC/action/storage_attestation","attest_author":"https://pith.science/pith/CQNJWC4XV6R5XAIMN47PTWQ4XC/action/author_attestation","sign_citation":"https://pith.science/pith/CQNJWC4XV6R5XAIMN47PTWQ4XC/action/citation_signature","submit_replication":"https://pith.science/pith/CQNJWC4XV6R5XAIMN47PTWQ4XC/action/replication_record"}},"created_at":"2026-07-05T00:27:35.753108+00:00","updated_at":"2026-07-05T00:27:35.753108+00:00"}