{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:TC2HAAPWIIVXLEP7HZRAOEOAVR","short_pith_number":"pith:TC2HAAPW","schema_version":"1.0","canonical_sha256":"98b47001f6422b7591ff3e620711c0ac49a1ba527cb09563aafa9bde06f078cc","source":{"kind":"arxiv","id":"2008.03979","version":2},"attestation_state":"computed","paper":{"title":"KR-BERT: A Small-Scale Korean-Specific Language Model","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hansol Jang, Hyopil Shin, Sangah Lee, Suzi Park, Yunmee Baik","submitted_at":"2020-08-10T09:26:00Z","abstract_excerpt":"Since the appearance of BERT, recent works including XLNet and RoBERTa utilize sentence embedding models pre-trained by large corpora and a large number of parameters. Because such models have large hardware and a huge amount of data, they take a long time to pre-train. Therefore it is important to attempt to make smaller models that perform comparatively. In this paper, we trained a Korean-specific model KR-BERT, utilizing a smaller vocabulary and dataset. Since Korean is one of the morphologically rich languages with poor resources using non-Latin alphabets, it is also important to capture l"},"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":"2008.03979","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-08-10T09:26:00Z","cross_cats_sorted":[],"title_canon_sha256":"7f70aef7a9bb0e4cf63bee640d8e68ec3a2ac94cf079ef0e1ea7fe6e334d10a4","abstract_canon_sha256":"60107ac2a124f580d68fa8c7ab9cb9882e1aba98e19f523bd3cbe663ef170ec6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:26:23.972036Z","signature_b64":"+MqRM3q+iHq06BoyzjQTRP+9YxnRqTerQ+vNrq4tEkmfMLEb6akf4jih3sq4QNGWqsDRt6BFf5Ad2DQGFKdFDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"98b47001f6422b7591ff3e620711c0ac49a1ba527cb09563aafa9bde06f078cc","last_reissued_at":"2026-07-05T01:26:23.971565Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:26:23.971565Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"KR-BERT: A Small-Scale Korean-Specific Language Model","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hansol Jang, Hyopil Shin, Sangah Lee, Suzi Park, Yunmee Baik","submitted_at":"2020-08-10T09:26:00Z","abstract_excerpt":"Since the appearance of BERT, recent works including XLNet and RoBERTa utilize sentence embedding models pre-trained by large corpora and a large number of parameters. Because such models have large hardware and a huge amount of data, they take a long time to pre-train. Therefore it is important to attempt to make smaller models that perform comparatively. In this paper, we trained a Korean-specific model KR-BERT, utilizing a smaller vocabulary and dataset. Since Korean is one of the morphologically rich languages with poor resources using non-Latin alphabets, it is also important to capture l"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2008.03979","kind":"arxiv","version":2},"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/2008.03979/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":"2008.03979","created_at":"2026-07-05T01:26:23.971644+00:00"},{"alias_kind":"arxiv_version","alias_value":"2008.03979v2","created_at":"2026-07-05T01:26:23.971644+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2008.03979","created_at":"2026-07-05T01:26:23.971644+00:00"},{"alias_kind":"pith_short_12","alias_value":"TC2HAAPWIIVX","created_at":"2026-07-05T01:26:23.971644+00:00"},{"alias_kind":"pith_short_16","alias_value":"TC2HAAPWIIVXLEP7","created_at":"2026-07-05T01:26:23.971644+00:00"},{"alias_kind":"pith_short_8","alias_value":"TC2HAAPW","created_at":"2026-07-05T01:26:23.971644+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.20089","citing_title":"IHUBERT: Vector-Based Semantic Deduplication and Domain-Balanced Pretraining for Persian Resources","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TC2HAAPWIIVXLEP7HZRAOEOAVR","json":"https://pith.science/pith/TC2HAAPWIIVXLEP7HZRAOEOAVR.json","graph_json":"https://pith.science/api/pith-number/TC2HAAPWIIVXLEP7HZRAOEOAVR/graph.json","events_json":"https://pith.science/api/pith-number/TC2HAAPWIIVXLEP7HZRAOEOAVR/events.json","paper":"https://pith.science/paper/TC2HAAPW"},"agent_actions":{"view_html":"https://pith.science/pith/TC2HAAPWIIVXLEP7HZRAOEOAVR","download_json":"https://pith.science/pith/TC2HAAPWIIVXLEP7HZRAOEOAVR.json","view_paper":"https://pith.science/paper/TC2HAAPW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2008.03979&json=true","fetch_graph":"https://pith.science/api/pith-number/TC2HAAPWIIVXLEP7HZRAOEOAVR/graph.json","fetch_events":"https://pith.science/api/pith-number/TC2HAAPWIIVXLEP7HZRAOEOAVR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TC2HAAPWIIVXLEP7HZRAOEOAVR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TC2HAAPWIIVXLEP7HZRAOEOAVR/action/storage_attestation","attest_author":"https://pith.science/pith/TC2HAAPWIIVXLEP7HZRAOEOAVR/action/author_attestation","sign_citation":"https://pith.science/pith/TC2HAAPWIIVXLEP7HZRAOEOAVR/action/citation_signature","submit_replication":"https://pith.science/pith/TC2HAAPWIIVXLEP7HZRAOEOAVR/action/replication_record"}},"created_at":"2026-07-05T01:26:23.971644+00:00","updated_at":"2026-07-05T01:26:23.971644+00:00"}