{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:KA3ZNVIHDUASGGUWF42HUTHC55","short_pith_number":"pith:KA3ZNVIH","schema_version":"1.0","canonical_sha256":"503796d5071d01231a962f347a4ce2ef6564778f93bf3d5224cc1c7fa6b16e68","source":{"kind":"arxiv","id":"2305.13707","version":1},"attestation_state":"computed","paper":{"title":"Do All Languages Cost the Same? Tokenization in the Era of Commercial Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"David R. Mortensen, Hila Gonen, Jungo Kasai, Noah A. Smith, Orevaoghene Ahia, Sachin Kumar, Yulia Tsvetkov","submitted_at":"2023-05-23T05:46:45Z","abstract_excerpt":"Language models have graduated from being research prototypes to commercialized products offered as web APIs, and recent works have highlighted the multilingual capabilities of these products. The API vendors charge their users based on usage, more specifically on the number of ``tokens'' processed or generated by the underlying language models. What constitutes a token, however, is training data and model dependent with a large variance in the number of tokens required to convey the same information in different languages. In this work, we analyze the effect of this non-uniformity on the fair"},"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":"2305.13707","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-23T05:46:45Z","cross_cats_sorted":[],"title_canon_sha256":"1e5180bf224a8ad67cfa5623f106515e334f5b8eb117eda8c50e3b591701b6af","abstract_canon_sha256":"7cbad57c87c21f8f36dafa1f66a221cff2a1a4e2d4ee6b1bbd74d2b312e2f125"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:12:56.453081Z","signature_b64":"lxn48X7wVNrCa5Et8p9WaNvM4K+8VYdcM3w+3IBropwyY/X68RpkzpwBmoVg4xsIFUoNdPHO1vc9AyvYipCLDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"503796d5071d01231a962f347a4ce2ef6564778f93bf3d5224cc1c7fa6b16e68","last_reissued_at":"2026-07-05T06:12:56.452660Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:12:56.452660Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Do All Languages Cost the Same? Tokenization in the Era of Commercial Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"David R. Mortensen, Hila Gonen, Jungo Kasai, Noah A. Smith, Orevaoghene Ahia, Sachin Kumar, Yulia Tsvetkov","submitted_at":"2023-05-23T05:46:45Z","abstract_excerpt":"Language models have graduated from being research prototypes to commercialized products offered as web APIs, and recent works have highlighted the multilingual capabilities of these products. The API vendors charge their users based on usage, more specifically on the number of ``tokens'' processed or generated by the underlying language models. What constitutes a token, however, is training data and model dependent with a large variance in the number of tokens required to convey the same information in different languages. In this work, we analyze the effect of this non-uniformity on the fair"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.13707","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/2305.13707/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":"2305.13707","created_at":"2026-07-05T06:12:56.452717+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.13707v1","created_at":"2026-07-05T06:12:56.452717+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.13707","created_at":"2026-07-05T06:12:56.452717+00:00"},{"alias_kind":"pith_short_12","alias_value":"KA3ZNVIHDUAS","created_at":"2026-07-05T06:12:56.452717+00:00"},{"alias_kind":"pith_short_16","alias_value":"KA3ZNVIHDUASGGUW","created_at":"2026-07-05T06:12:56.452717+00:00"},{"alias_kind":"pith_short_8","alias_value":"KA3ZNVIH","created_at":"2026-07-05T06:12:56.452717+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.20993","citing_title":"Phonemes to the Rescue: Multilingual Tokenization Based on International Phonetic Alphabet","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2605.24718","citing_title":"The Tokenizer Tax Across 25 European Languages: Domain Invariance, Cross-Lingual Few-Shot Effects, and the Ukrainian Penalty","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2605.29379","citing_title":"BrahmicTokenizer-131K: An Indic-Capable Drop-In Replacement for o200k_base","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2604.02492","citing_title":"Token-Efficient Multimodal Reasoning via Image Prompt Packaging","ref_index":11,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KA3ZNVIHDUASGGUWF42HUTHC55","json":"https://pith.science/pith/KA3ZNVIHDUASGGUWF42HUTHC55.json","graph_json":"https://pith.science/api/pith-number/KA3ZNVIHDUASGGUWF42HUTHC55/graph.json","events_json":"https://pith.science/api/pith-number/KA3ZNVIHDUASGGUWF42HUTHC55/events.json","paper":"https://pith.science/paper/KA3ZNVIH"},"agent_actions":{"view_html":"https://pith.science/pith/KA3ZNVIHDUASGGUWF42HUTHC55","download_json":"https://pith.science/pith/KA3ZNVIHDUASGGUWF42HUTHC55.json","view_paper":"https://pith.science/paper/KA3ZNVIH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.13707&json=true","fetch_graph":"https://pith.science/api/pith-number/KA3ZNVIHDUASGGUWF42HUTHC55/graph.json","fetch_events":"https://pith.science/api/pith-number/KA3ZNVIHDUASGGUWF42HUTHC55/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KA3ZNVIHDUASGGUWF42HUTHC55/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KA3ZNVIHDUASGGUWF42HUTHC55/action/storage_attestation","attest_author":"https://pith.science/pith/KA3ZNVIHDUASGGUWF42HUTHC55/action/author_attestation","sign_citation":"https://pith.science/pith/KA3ZNVIHDUASGGUWF42HUTHC55/action/citation_signature","submit_replication":"https://pith.science/pith/KA3ZNVIHDUASGGUWF42HUTHC55/action/replication_record"}},"created_at":"2026-07-05T06:12:56.452717+00:00","updated_at":"2026-07-05T06:12:56.452717+00:00"}