{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:3XBVZCJQ32HMYJ77FS4MB2CA6P","short_pith_number":"pith:3XBVZCJQ","schema_version":"1.0","canonical_sha256":"ddc35c8930de8ecc27ff2cb8c0e840f3e644d3ebe1415e75b04fd068f8ce79a0","source":{"kind":"arxiv","id":"2310.03477","version":1},"attestation_state":"computed","paper":{"title":"Tik-to-Tok: Translating Language Models One Token at a Time: An Embedding Initialization Strategy for Efficient Language Adaptation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bettina Berendt, Fran\\c{c}ois Remy, Kris Demuynck, Pieter Delobelle, Thomas Demeester","submitted_at":"2023-10-05T11:45:29Z","abstract_excerpt":"Training monolingual language models for low and mid-resource languages is made challenging by limited and often inadequate pretraining data. In this study, we propose a novel model conversion strategy to address this issue, adapting high-resources monolingual language models to a new target language. By generalizing over a word translation dictionary encompassing both the source and target languages, we map tokens from the target tokenizer to semantically similar tokens from the source language tokenizer. This one-to-many token mapping improves tremendously the initialization of the embedding"},"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":"2310.03477","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-10-05T11:45:29Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"7357d85abc734d22d0cf6501843a07419a9d66c84ca0d5753ce68451d4dec08c","abstract_canon_sha256":"64059e596c7871c30fa0b7deef606e5b8257d5f2de05dc7aa110cf5377ecaa1c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:57:34.139900Z","signature_b64":"M7xCgfHvA9jRoXyLBGo6w4hLOewrZp7cXgM5gOy/yAgCBg7LkgJtcCSMKvXDZxwlf5xANuotwo9l33h7MNxvAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ddc35c8930de8ecc27ff2cb8c0e840f3e644d3ebe1415e75b04fd068f8ce79a0","last_reissued_at":"2026-07-05T06:57:34.139262Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:57:34.139262Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Tik-to-Tok: Translating Language Models One Token at a Time: An Embedding Initialization Strategy for Efficient Language Adaptation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bettina Berendt, Fran\\c{c}ois Remy, Kris Demuynck, Pieter Delobelle, Thomas Demeester","submitted_at":"2023-10-05T11:45:29Z","abstract_excerpt":"Training monolingual language models for low and mid-resource languages is made challenging by limited and often inadequate pretraining data. In this study, we propose a novel model conversion strategy to address this issue, adapting high-resources monolingual language models to a new target language. By generalizing over a word translation dictionary encompassing both the source and target languages, we map tokens from the target tokenizer to semantically similar tokens from the source language tokenizer. This one-to-many token mapping improves tremendously the initialization of the embedding"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.03477","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/2310.03477/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":"2310.03477","created_at":"2026-07-05T06:57:34.139322+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.03477v1","created_at":"2026-07-05T06:57:34.139322+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.03477","created_at":"2026-07-05T06:57:34.139322+00:00"},{"alias_kind":"pith_short_12","alias_value":"3XBVZCJQ32HM","created_at":"2026-07-05T06:57:34.139322+00:00"},{"alias_kind":"pith_short_16","alias_value":"3XBVZCJQ32HMYJ77","created_at":"2026-07-05T06:57:34.139322+00:00"},{"alias_kind":"pith_short_8","alias_value":"3XBVZCJQ","created_at":"2026-07-05T06:57:34.139322+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2608.00582","citing_title":"Writing-System-Level Tokenizer Adaptation for Byte-Level BPE","ref_index":2019,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3XBVZCJQ32HMYJ77FS4MB2CA6P","json":"https://pith.science/pith/3XBVZCJQ32HMYJ77FS4MB2CA6P.json","graph_json":"https://pith.science/api/pith-number/3XBVZCJQ32HMYJ77FS4MB2CA6P/graph.json","events_json":"https://pith.science/api/pith-number/3XBVZCJQ32HMYJ77FS4MB2CA6P/events.json","paper":"https://pith.science/paper/3XBVZCJQ"},"agent_actions":{"view_html":"https://pith.science/pith/3XBVZCJQ32HMYJ77FS4MB2CA6P","download_json":"https://pith.science/pith/3XBVZCJQ32HMYJ77FS4MB2CA6P.json","view_paper":"https://pith.science/paper/3XBVZCJQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.03477&json=true","fetch_graph":"https://pith.science/api/pith-number/3XBVZCJQ32HMYJ77FS4MB2CA6P/graph.json","fetch_events":"https://pith.science/api/pith-number/3XBVZCJQ32HMYJ77FS4MB2CA6P/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3XBVZCJQ32HMYJ77FS4MB2CA6P/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3XBVZCJQ32HMYJ77FS4MB2CA6P/action/storage_attestation","attest_author":"https://pith.science/pith/3XBVZCJQ32HMYJ77FS4MB2CA6P/action/author_attestation","sign_citation":"https://pith.science/pith/3XBVZCJQ32HMYJ77FS4MB2CA6P/action/citation_signature","submit_replication":"https://pith.science/pith/3XBVZCJQ32HMYJ77FS4MB2CA6P/action/replication_record"}},"created_at":"2026-07-05T06:57:34.139322+00:00","updated_at":"2026-07-05T06:57:34.139322+00:00"}