{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:OQK2PIPXZP33FZDJ4MMHHCNEFS","short_pith_number":"pith:OQK2PIPX","schema_version":"1.0","canonical_sha256":"7415a7a1f7cbf7b2e469e3187389a42cb64cc0a6c5e58bd1fbf24fa5d31d0a92","source":{"kind":"arxiv","id":"2403.10691","version":2},"attestation_state":"computed","paper":{"title":"MYTE: Morphology-Driven Byte Encoding for Better and Fairer Multilingual Language Modeling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Hila Gonen, Luke Zettlemoyer, Orevaoghene Ahia, Terra Blevins, Tomasz Limisiewicz","submitted_at":"2024-03-15T21:21:11Z","abstract_excerpt":"A major consideration in multilingual language modeling is how to best represent languages with diverse vocabularies and scripts. Although contemporary text encoding methods cover most of the world's writing systems, they exhibit bias towards the high-resource languages of the Global West. As a result, texts of underrepresented languages tend to be segmented into long sequences of linguistically meaningless units. To address the disparities, we introduce a new paradigm that encodes the same information with segments of consistent size across diverse languages. Our encoding convention (MYTE) is"},"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":"2403.10691","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-15T21:21:11Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"06ff7c9c3095b8b9cb906173d7d9bf0f38969039e98913fa06972d6a5b098a25","abstract_canon_sha256":"f1151159cdb1f414bcc506a6b5b0758a3194b12b6c2c0cb660b2e5a458ae0a23"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:33:29.823765Z","signature_b64":"igE6ZOLFMsVRq9+zh7KAsfC6R4hEs1ZK0PIKKBUstBJpB1+6HQun90Ct3hfy+UIrPofGdGzzLuDomd95/7Z7BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7415a7a1f7cbf7b2e469e3187389a42cb64cc0a6c5e58bd1fbf24fa5d31d0a92","last_reissued_at":"2026-07-05T09:33:29.823290Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:33:29.823290Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MYTE: Morphology-Driven Byte Encoding for Better and Fairer Multilingual Language Modeling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Hila Gonen, Luke Zettlemoyer, Orevaoghene Ahia, Terra Blevins, Tomasz Limisiewicz","submitted_at":"2024-03-15T21:21:11Z","abstract_excerpt":"A major consideration in multilingual language modeling is how to best represent languages with diverse vocabularies and scripts. Although contemporary text encoding methods cover most of the world's writing systems, they exhibit bias towards the high-resource languages of the Global West. As a result, texts of underrepresented languages tend to be segmented into long sequences of linguistically meaningless units. To address the disparities, we introduce a new paradigm that encodes the same information with segments of consistent size across diverse languages. Our encoding convention (MYTE) is"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.10691","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/2403.10691/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":"2403.10691","created_at":"2026-07-05T09:33:29.823360+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.10691v2","created_at":"2026-07-05T09:33:29.823360+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.10691","created_at":"2026-07-05T09:33:29.823360+00:00"},{"alias_kind":"pith_short_12","alias_value":"OQK2PIPXZP33","created_at":"2026-07-05T09:33:29.823360+00:00"},{"alias_kind":"pith_short_16","alias_value":"OQK2PIPXZP33FZDJ","created_at":"2026-07-05T09:33:29.823360+00:00"},{"alias_kind":"pith_short_8","alias_value":"OQK2PIPX","created_at":"2026-07-05T09:33:29.823360+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2601.07220","citing_title":"The Roots of Performance Disparity in Multilingual Language Models: Intrinsic Modeling Difficulty or Design Choices?","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09630","citing_title":"Scratchpad Patching: Decoupling Compute from Patch Size in Byte-Level Language Models","ref_index":58,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OQK2PIPXZP33FZDJ4MMHHCNEFS","json":"https://pith.science/pith/OQK2PIPXZP33FZDJ4MMHHCNEFS.json","graph_json":"https://pith.science/api/pith-number/OQK2PIPXZP33FZDJ4MMHHCNEFS/graph.json","events_json":"https://pith.science/api/pith-number/OQK2PIPXZP33FZDJ4MMHHCNEFS/events.json","paper":"https://pith.science/paper/OQK2PIPX"},"agent_actions":{"view_html":"https://pith.science/pith/OQK2PIPXZP33FZDJ4MMHHCNEFS","download_json":"https://pith.science/pith/OQK2PIPXZP33FZDJ4MMHHCNEFS.json","view_paper":"https://pith.science/paper/OQK2PIPX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.10691&json=true","fetch_graph":"https://pith.science/api/pith-number/OQK2PIPXZP33FZDJ4MMHHCNEFS/graph.json","fetch_events":"https://pith.science/api/pith-number/OQK2PIPXZP33FZDJ4MMHHCNEFS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OQK2PIPXZP33FZDJ4MMHHCNEFS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OQK2PIPXZP33FZDJ4MMHHCNEFS/action/storage_attestation","attest_author":"https://pith.science/pith/OQK2PIPXZP33FZDJ4MMHHCNEFS/action/author_attestation","sign_citation":"https://pith.science/pith/OQK2PIPXZP33FZDJ4MMHHCNEFS/action/citation_signature","submit_replication":"https://pith.science/pith/OQK2PIPXZP33FZDJ4MMHHCNEFS/action/replication_record"}},"created_at":"2026-07-05T09:33:29.823360+00:00","updated_at":"2026-07-05T09:33:29.823360+00:00"}