{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:2P4NKKFRHWO5QAXZYHRVIOXATW","short_pith_number":"pith:2P4NKKFR","schema_version":"1.0","canonical_sha256":"d3f8d528b13d9dd802f9c1e3543ae09dac10dadb44fc3ee21fca52d9330b8054","source":{"kind":"arxiv","id":"2504.17025","version":1},"attestation_state":"computed","paper":{"title":"Optimizing LLMs for Italian: Reducing Token Fertility and Enhancing Efficiency Through Vocabulary Adaptation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alessio Miaschi, Andrea Esuli, Andrei Stefan Bejgu, Edoardo Barba, Felice Dell'Orletta, Giovanni Puccetti, Luca Moroni, Pere-Lluis Huguet Cabot, Roberto Navigli","submitted_at":"2025-04-23T18:12:27Z","abstract_excerpt":"The number of pretrained Large Language Models (LLMs) is increasing steadily, though the majority are designed predominantly for the English language. While state-of-the-art LLMs can handle other languages, due to language contamination or some degree of multilingual pretraining data, they are not optimized for non-English languages, leading to inefficient encoding (high token \"fertility\") and slower inference speed. In this work, we thoroughly compare a variety of vocabulary adaptation techniques for optimizing English LLMs for the Italian language, and put forward Semantic Alignment Vocabula"},"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":"2504.17025","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-23T18:12:27Z","cross_cats_sorted":[],"title_canon_sha256":"ad20e0efff760c7fb5c08caf5695aaacc399ed27a9f64b566f8260ee15dd60e8","abstract_canon_sha256":"a97c6909c87fa8c958cb6aae06eee7d02f1af30d6d057c10a050c0437bd6de10"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:53:20.771101Z","signature_b64":"pooU9pWde6Sc/PjJeuNTPSDVfZ5uRwMaJauKG66yoOfYVGdy9LQJSoFiSEEtIRs4OF+8+7RmlQsnKELePAVcAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d3f8d528b13d9dd802f9c1e3543ae09dac10dadb44fc3ee21fca52d9330b8054","last_reissued_at":"2026-07-05T10:53:20.770600Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:53:20.770600Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Optimizing LLMs for Italian: Reducing Token Fertility and Enhancing Efficiency Through Vocabulary Adaptation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alessio Miaschi, Andrea Esuli, Andrei Stefan Bejgu, Edoardo Barba, Felice Dell'Orletta, Giovanni Puccetti, Luca Moroni, Pere-Lluis Huguet Cabot, Roberto Navigli","submitted_at":"2025-04-23T18:12:27Z","abstract_excerpt":"The number of pretrained Large Language Models (LLMs) is increasing steadily, though the majority are designed predominantly for the English language. While state-of-the-art LLMs can handle other languages, due to language contamination or some degree of multilingual pretraining data, they are not optimized for non-English languages, leading to inefficient encoding (high token \"fertility\") and slower inference speed. In this work, we thoroughly compare a variety of vocabulary adaptation techniques for optimizing English LLMs for the Italian language, and put forward Semantic Alignment Vocabula"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.17025","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/2504.17025/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":"2504.17025","created_at":"2026-07-05T10:53:20.770662+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.17025v1","created_at":"2026-07-05T10:53:20.770662+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.17025","created_at":"2026-07-05T10:53:20.770662+00:00"},{"alias_kind":"pith_short_12","alias_value":"2P4NKKFRHWO5","created_at":"2026-07-05T10:53:20.770662+00:00"},{"alias_kind":"pith_short_16","alias_value":"2P4NKKFRHWO5QAXZ","created_at":"2026-07-05T10:53:20.770662+00:00"},{"alias_kind":"pith_short_8","alias_value":"2P4NKKFR","created_at":"2026-07-05T10:53:20.770662+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/2P4NKKFRHWO5QAXZYHRVIOXATW","json":"https://pith.science/pith/2P4NKKFRHWO5QAXZYHRVIOXATW.json","graph_json":"https://pith.science/api/pith-number/2P4NKKFRHWO5QAXZYHRVIOXATW/graph.json","events_json":"https://pith.science/api/pith-number/2P4NKKFRHWO5QAXZYHRVIOXATW/events.json","paper":"https://pith.science/paper/2P4NKKFR"},"agent_actions":{"view_html":"https://pith.science/pith/2P4NKKFRHWO5QAXZYHRVIOXATW","download_json":"https://pith.science/pith/2P4NKKFRHWO5QAXZYHRVIOXATW.json","view_paper":"https://pith.science/paper/2P4NKKFR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.17025&json=true","fetch_graph":"https://pith.science/api/pith-number/2P4NKKFRHWO5QAXZYHRVIOXATW/graph.json","fetch_events":"https://pith.science/api/pith-number/2P4NKKFRHWO5QAXZYHRVIOXATW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2P4NKKFRHWO5QAXZYHRVIOXATW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2P4NKKFRHWO5QAXZYHRVIOXATW/action/storage_attestation","attest_author":"https://pith.science/pith/2P4NKKFRHWO5QAXZYHRVIOXATW/action/author_attestation","sign_citation":"https://pith.science/pith/2P4NKKFRHWO5QAXZYHRVIOXATW/action/citation_signature","submit_replication":"https://pith.science/pith/2P4NKKFRHWO5QAXZYHRVIOXATW/action/replication_record"}},"created_at":"2026-07-05T10:53:20.770662+00:00","updated_at":"2026-07-05T10:53:20.770662+00:00"}