{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:V4AU7NUPW5ZVIZAR6NNRKG2WLN","short_pith_number":"pith:V4AU7NUP","schema_version":"1.0","canonical_sha256":"af014fb68fb773546411f35b151b565b623a01e0f95cd7c852b061072da67921","source":{"kind":"arxiv","id":"2506.17789","version":3},"attestation_state":"computed","paper":{"title":"Multilingual Tokenization through the Lens of Indian Languages: Challenges and Insights","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ganesh Ramakrishnan, Maharaj Brahma, Maunendra Sankar Desarkar, Nagasai Saketh Naidu, N J Karthika, Rajat Verma, Rohit Saluja","submitted_at":"2025-06-21T18:47:33Z","abstract_excerpt":"Tokenization plays a pivotal role in NLP and is fundamental to training language models. However, existing tokenizers are often skewed towards high-resource languages, limiting their effectiveness for linguistically diverse and morphologically rich languages such as those in the Indian subcontinent. In this work, we present a comprehensive empirical study of multilingual tokenization across 17 Indic languages spanning 11 scripts and two language families. We systematically evaluate the effects of (i) widely used subword algorithms: BPE and Unigram LM, (ii) script and orthography-aware normaliz"},"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":"2506.17789","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-21T18:47:33Z","cross_cats_sorted":[],"title_canon_sha256":"26f74f5f266edc3d20a9ceae087cba90b6c09cab4d72a59d1c332d9cda8e05bc","abstract_canon_sha256":"2042bcad5ab29f2e5233549751c8150614cb628e70321c4daf308e5cf9c5d82a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T01:11:56.026699Z","signature_b64":"OrUQLAi1i+3jLGXEpR++f6itHDEFqBCGCsiJNd/5eOA3sITaPlJ6Us+EhGuUh/ShMm7g7zUBy76MPhceDJa4BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"af014fb68fb773546411f35b151b565b623a01e0f95cd7c852b061072da67921","last_reissued_at":"2026-06-23T01:11:56.026025Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T01:11:56.026025Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multilingual Tokenization through the Lens of Indian Languages: Challenges and Insights","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ganesh Ramakrishnan, Maharaj Brahma, Maunendra Sankar Desarkar, Nagasai Saketh Naidu, N J Karthika, Rajat Verma, Rohit Saluja","submitted_at":"2025-06-21T18:47:33Z","abstract_excerpt":"Tokenization plays a pivotal role in NLP and is fundamental to training language models. However, existing tokenizers are often skewed towards high-resource languages, limiting their effectiveness for linguistically diverse and morphologically rich languages such as those in the Indian subcontinent. In this work, we present a comprehensive empirical study of multilingual tokenization across 17 Indic languages spanning 11 scripts and two language families. We systematically evaluate the effects of (i) widely used subword algorithms: BPE and Unigram LM, (ii) script and orthography-aware normaliz"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.17789","kind":"arxiv","version":3},"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/2506.17789/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":"2506.17789","created_at":"2026-06-23T01:11:56.026118+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.17789v3","created_at":"2026-06-23T01:11:56.026118+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.17789","created_at":"2026-06-23T01:11:56.026118+00:00"},{"alias_kind":"pith_short_12","alias_value":"V4AU7NUPW5ZV","created_at":"2026-06-23T01:11:56.026118+00:00"},{"alias_kind":"pith_short_16","alias_value":"V4AU7NUPW5ZVIZAR","created_at":"2026-06-23T01:11:56.026118+00:00"},{"alias_kind":"pith_short_8","alias_value":"V4AU7NUP","created_at":"2026-06-23T01:11:56.026118+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/V4AU7NUPW5ZVIZAR6NNRKG2WLN","json":"https://pith.science/pith/V4AU7NUPW5ZVIZAR6NNRKG2WLN.json","graph_json":"https://pith.science/api/pith-number/V4AU7NUPW5ZVIZAR6NNRKG2WLN/graph.json","events_json":"https://pith.science/api/pith-number/V4AU7NUPW5ZVIZAR6NNRKG2WLN/events.json","paper":"https://pith.science/paper/V4AU7NUP"},"agent_actions":{"view_html":"https://pith.science/pith/V4AU7NUPW5ZVIZAR6NNRKG2WLN","download_json":"https://pith.science/pith/V4AU7NUPW5ZVIZAR6NNRKG2WLN.json","view_paper":"https://pith.science/paper/V4AU7NUP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.17789&json=true","fetch_graph":"https://pith.science/api/pith-number/V4AU7NUPW5ZVIZAR6NNRKG2WLN/graph.json","fetch_events":"https://pith.science/api/pith-number/V4AU7NUPW5ZVIZAR6NNRKG2WLN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/V4AU7NUPW5ZVIZAR6NNRKG2WLN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/V4AU7NUPW5ZVIZAR6NNRKG2WLN/action/storage_attestation","attest_author":"https://pith.science/pith/V4AU7NUPW5ZVIZAR6NNRKG2WLN/action/author_attestation","sign_citation":"https://pith.science/pith/V4AU7NUPW5ZVIZAR6NNRKG2WLN/action/citation_signature","submit_replication":"https://pith.science/pith/V4AU7NUPW5ZVIZAR6NNRKG2WLN/action/replication_record"}},"created_at":"2026-06-23T01:11:56.026118+00:00","updated_at":"2026-06-23T01:11:56.026118+00:00"}