{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:BL2TCVWWOTZXQXQNUHIPRLWOBT","short_pith_number":"pith:BL2TCVWW","schema_version":"1.0","canonical_sha256":"0af53156d674f3785e0da1d0f8aece0cdd58525fbb8ea58e968cf3236f18d69e","source":{"kind":"arxiv","id":"2402.09949","version":2},"attestation_state":"computed","paper":{"title":"Multi-word Tokenization for Sequence Compression","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Andrea Zugarini, Leonardo Rigutini, Leonidas Gee, Marco Ernandes","submitted_at":"2024-02-15T13:52:23Z","abstract_excerpt":"Large Language Models have proven highly successful at modelling a variety of tasks. However, this comes at a steep computational cost that hinders wider industrial uptake. In this paper, we present MWT: a Multi-Word Tokenizer that goes beyond word boundaries by representing frequent multi-word expressions as single tokens. MWTs produce a more compact and efficient tokenization that yields two benefits: (1) Increase in performance due to a greater coverage of input data given a fixed sequence length budget; (2) Faster and lighter inference due to the ability to reduce the sequence length with "},"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":"2402.09949","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-15T13:52:23Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"4b4d3c1fe988acb17a411b6138f74d7f637ddef4130686978e5293604c97dfe8","abstract_canon_sha256":"b4593776e5d66d3859634ca88f67b920155941ab5c48e8c6ba6191bd32953698"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:04:37.674331Z","signature_b64":"91SjhT8mTAFOSQ5/vpW0Assrwo653hVXcSWV/oGmr6z4PCQRH6269+VLQpAhQkC0Lcy387/f1/WCkwgERw2YCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0af53156d674f3785e0da1d0f8aece0cdd58525fbb8ea58e968cf3236f18d69e","last_reissued_at":"2026-07-05T08:04:37.673870Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:04:37.673870Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multi-word Tokenization for Sequence Compression","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Andrea Zugarini, Leonardo Rigutini, Leonidas Gee, Marco Ernandes","submitted_at":"2024-02-15T13:52:23Z","abstract_excerpt":"Large Language Models have proven highly successful at modelling a variety of tasks. However, this comes at a steep computational cost that hinders wider industrial uptake. In this paper, we present MWT: a Multi-Word Tokenizer that goes beyond word boundaries by representing frequent multi-word expressions as single tokens. MWTs produce a more compact and efficient tokenization that yields two benefits: (1) Increase in performance due to a greater coverage of input data given a fixed sequence length budget; (2) Faster and lighter inference due to the ability to reduce the sequence length with "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.09949","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/2402.09949/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":"2402.09949","created_at":"2026-07-05T08:04:37.673926+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.09949v2","created_at":"2026-07-05T08:04:37.673926+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.09949","created_at":"2026-07-05T08:04:37.673926+00:00"},{"alias_kind":"pith_short_12","alias_value":"BL2TCVWWOTZX","created_at":"2026-07-05T08:04:37.673926+00:00"},{"alias_kind":"pith_short_16","alias_value":"BL2TCVWWOTZXQXQN","created_at":"2026-07-05T08:04:37.673926+00:00"},{"alias_kind":"pith_short_8","alias_value":"BL2TCVWW","created_at":"2026-07-05T08:04:37.673926+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/BL2TCVWWOTZXQXQNUHIPRLWOBT","json":"https://pith.science/pith/BL2TCVWWOTZXQXQNUHIPRLWOBT.json","graph_json":"https://pith.science/api/pith-number/BL2TCVWWOTZXQXQNUHIPRLWOBT/graph.json","events_json":"https://pith.science/api/pith-number/BL2TCVWWOTZXQXQNUHIPRLWOBT/events.json","paper":"https://pith.science/paper/BL2TCVWW"},"agent_actions":{"view_html":"https://pith.science/pith/BL2TCVWWOTZXQXQNUHIPRLWOBT","download_json":"https://pith.science/pith/BL2TCVWWOTZXQXQNUHIPRLWOBT.json","view_paper":"https://pith.science/paper/BL2TCVWW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.09949&json=true","fetch_graph":"https://pith.science/api/pith-number/BL2TCVWWOTZXQXQNUHIPRLWOBT/graph.json","fetch_events":"https://pith.science/api/pith-number/BL2TCVWWOTZXQXQNUHIPRLWOBT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BL2TCVWWOTZXQXQNUHIPRLWOBT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BL2TCVWWOTZXQXQNUHIPRLWOBT/action/storage_attestation","attest_author":"https://pith.science/pith/BL2TCVWWOTZXQXQNUHIPRLWOBT/action/author_attestation","sign_citation":"https://pith.science/pith/BL2TCVWWOTZXQXQNUHIPRLWOBT/action/citation_signature","submit_replication":"https://pith.science/pith/BL2TCVWWOTZXQXQNUHIPRLWOBT/action/replication_record"}},"created_at":"2026-07-05T08:04:37.673926+00:00","updated_at":"2026-07-05T08:04:37.673926+00:00"}