{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:5OJD4KDBUHE7UTRQKBD4YGILXJ","short_pith_number":"pith:5OJD4KDB","schema_version":"1.0","canonical_sha256":"eb923e2861a1c9fa4e305047cc190bba5f97d39330367752640a07a576edf37c","source":{"kind":"arxiv","id":"2506.15138","version":1},"attestation_state":"computed","paper":{"title":"Thunder-Tok: Minimizing Tokens per Word in Tokenizing Korean Texts for Generative Language Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chanwoo Park, Gyeongje Cho, Jaejin Lee, Sangmin Lee, Sungmok Jung, Yeonkyoun So","submitted_at":"2025-06-18T04:40:44Z","abstract_excerpt":"This paper introduces Thunder-Tok, a new Korean tokenizer designed to reduce token fertility without compromising model performance. Our approach uses a rule-based pre-tokenization method that aligns with the linguistic structure of the Korean language. We also create a seed vocabulary containing tokens that resemble linguistic units and employ a branching entropy-based selection algorithm. These techniques increase the average token length, thus lowering fertility while preserving linguistic information. Experimental results indicate that Thunder-Tok reduces fertility by approximately 10% (i."},"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":"2506.15138","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-18T04:40:44Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2396105d68bfb42ab838a96a11f6e392da0d3aae0fd87869f199058e8bd9b681","abstract_canon_sha256":"c3ee9d65a03c7c384b3a53986f28ddb3e3aeb3aa26c01bf9f93139e5938eef16"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:23:37.817626Z","signature_b64":"M38yooqFODWGK65i55o/U9VwfKyIdJOsVGgt00mSkj/Vf1ucc9bqvetlYYqH6Yid8AQvm9TE4bg8aT4spNSiAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"eb923e2861a1c9fa4e305047cc190bba5f97d39330367752640a07a576edf37c","last_reissued_at":"2026-07-05T11:23:37.817072Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:23:37.817072Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Thunder-Tok: Minimizing Tokens per Word in Tokenizing Korean Texts for Generative Language Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chanwoo Park, Gyeongje Cho, Jaejin Lee, Sangmin Lee, Sungmok Jung, Yeonkyoun So","submitted_at":"2025-06-18T04:40:44Z","abstract_excerpt":"This paper introduces Thunder-Tok, a new Korean tokenizer designed to reduce token fertility without compromising model performance. Our approach uses a rule-based pre-tokenization method that aligns with the linguistic structure of the Korean language. We also create a seed vocabulary containing tokens that resemble linguistic units and employ a branching entropy-based selection algorithm. These techniques increase the average token length, thus lowering fertility while preserving linguistic information. Experimental results indicate that Thunder-Tok reduces fertility by approximately 10% (i."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.15138","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/2506.15138/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":"2506.15138","created_at":"2026-07-05T11:23:37.817131+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.15138v1","created_at":"2026-07-05T11:23:37.817131+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.15138","created_at":"2026-07-05T11:23:37.817131+00:00"},{"alias_kind":"pith_short_12","alias_value":"5OJD4KDBUHE7","created_at":"2026-07-05T11:23:37.817131+00:00"},{"alias_kind":"pith_short_16","alias_value":"5OJD4KDBUHE7UTRQ","created_at":"2026-07-05T11:23:37.817131+00:00"},{"alias_kind":"pith_short_8","alias_value":"5OJD4KDB","created_at":"2026-07-05T11:23:37.817131+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/5OJD4KDBUHE7UTRQKBD4YGILXJ","json":"https://pith.science/pith/5OJD4KDBUHE7UTRQKBD4YGILXJ.json","graph_json":"https://pith.science/api/pith-number/5OJD4KDBUHE7UTRQKBD4YGILXJ/graph.json","events_json":"https://pith.science/api/pith-number/5OJD4KDBUHE7UTRQKBD4YGILXJ/events.json","paper":"https://pith.science/paper/5OJD4KDB"},"agent_actions":{"view_html":"https://pith.science/pith/5OJD4KDBUHE7UTRQKBD4YGILXJ","download_json":"https://pith.science/pith/5OJD4KDBUHE7UTRQKBD4YGILXJ.json","view_paper":"https://pith.science/paper/5OJD4KDB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.15138&json=true","fetch_graph":"https://pith.science/api/pith-number/5OJD4KDBUHE7UTRQKBD4YGILXJ/graph.json","fetch_events":"https://pith.science/api/pith-number/5OJD4KDBUHE7UTRQKBD4YGILXJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5OJD4KDBUHE7UTRQKBD4YGILXJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5OJD4KDBUHE7UTRQKBD4YGILXJ/action/storage_attestation","attest_author":"https://pith.science/pith/5OJD4KDBUHE7UTRQKBD4YGILXJ/action/author_attestation","sign_citation":"https://pith.science/pith/5OJD4KDBUHE7UTRQKBD4YGILXJ/action/citation_signature","submit_replication":"https://pith.science/pith/5OJD4KDBUHE7UTRQKBD4YGILXJ/action/replication_record"}},"created_at":"2026-07-05T11:23:37.817131+00:00","updated_at":"2026-07-05T11:23:37.817131+00:00"}