{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:FHE4BGGYIUITGQNNXURI7LPTKJ","short_pith_number":"pith:FHE4BGGY","schema_version":"1.0","canonical_sha256":"29c9c098d845113341adbd228fadf3527191d02ad2b4d6896462de07c9375958","source":{"kind":"arxiv","id":"2508.15190","version":1},"attestation_state":"computed","paper":{"title":"SemToken: Semantic-Aware Tokenization for Efficient Long-Context Language Modeling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Dong Liu, Yanxuan Yu","submitted_at":"2025-08-21T03:01:53Z","abstract_excerpt":"Tokenization plays a critical role in language modeling, yet existing approaches such as Byte-Pair Encoding (BPE) or WordPiece operate purely on frequency statistics, ignoring the underlying semantic structure of text. This leads to over-tokenization of semantically redundant spans and underutilization of contextual coherence, particularly in long-context scenarios. In this work, we propose \\textbf{SemToken}, a semantic-aware tokenization framework that jointly reduces token redundancy and improves computation efficiency. SemToken first extracts contextual semantic embeddings via lightweight e"},"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":"2508.15190","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-21T03:01:53Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"93c095879ffba5c50717fa051dc3b9093354ab005b75249b1913f1fdf2af8f5b","abstract_canon_sha256":"f9e15ed6e76e5ad40d917349b6f075f20062029ef5c359f27145f424b7eeaba0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:57:07.997567Z","signature_b64":"Glq2/aDGhItfvIfj8SCxl4kQu1pkIFmFY+gHogbPvyq/UEL6xVW2vi76QB4oeECDa/LanDutCW/uoocqWNe+Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"29c9c098d845113341adbd228fadf3527191d02ad2b4d6896462de07c9375958","last_reissued_at":"2026-07-05T11:57:07.997150Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:57:07.997150Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SemToken: Semantic-Aware Tokenization for Efficient Long-Context Language Modeling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Dong Liu, Yanxuan Yu","submitted_at":"2025-08-21T03:01:53Z","abstract_excerpt":"Tokenization plays a critical role in language modeling, yet existing approaches such as Byte-Pair Encoding (BPE) or WordPiece operate purely on frequency statistics, ignoring the underlying semantic structure of text. This leads to over-tokenization of semantically redundant spans and underutilization of contextual coherence, particularly in long-context scenarios. In this work, we propose \\textbf{SemToken}, a semantic-aware tokenization framework that jointly reduces token redundancy and improves computation efficiency. SemToken first extracts contextual semantic embeddings via lightweight e"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.15190","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/2508.15190/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":"2508.15190","created_at":"2026-07-05T11:57:07.997203+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.15190v1","created_at":"2026-07-05T11:57:07.997203+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.15190","created_at":"2026-07-05T11:57:07.997203+00:00"},{"alias_kind":"pith_short_12","alias_value":"FHE4BGGYIUIT","created_at":"2026-07-05T11:57:07.997203+00:00"},{"alias_kind":"pith_short_16","alias_value":"FHE4BGGYIUITGQNN","created_at":"2026-07-05T11:57:07.997203+00:00"},{"alias_kind":"pith_short_8","alias_value":"FHE4BGGY","created_at":"2026-07-05T11:57:07.997203+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/FHE4BGGYIUITGQNNXURI7LPTKJ","json":"https://pith.science/pith/FHE4BGGYIUITGQNNXURI7LPTKJ.json","graph_json":"https://pith.science/api/pith-number/FHE4BGGYIUITGQNNXURI7LPTKJ/graph.json","events_json":"https://pith.science/api/pith-number/FHE4BGGYIUITGQNNXURI7LPTKJ/events.json","paper":"https://pith.science/paper/FHE4BGGY"},"agent_actions":{"view_html":"https://pith.science/pith/FHE4BGGYIUITGQNNXURI7LPTKJ","download_json":"https://pith.science/pith/FHE4BGGYIUITGQNNXURI7LPTKJ.json","view_paper":"https://pith.science/paper/FHE4BGGY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.15190&json=true","fetch_graph":"https://pith.science/api/pith-number/FHE4BGGYIUITGQNNXURI7LPTKJ/graph.json","fetch_events":"https://pith.science/api/pith-number/FHE4BGGYIUITGQNNXURI7LPTKJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FHE4BGGYIUITGQNNXURI7LPTKJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FHE4BGGYIUITGQNNXURI7LPTKJ/action/storage_attestation","attest_author":"https://pith.science/pith/FHE4BGGYIUITGQNNXURI7LPTKJ/action/author_attestation","sign_citation":"https://pith.science/pith/FHE4BGGYIUITGQNNXURI7LPTKJ/action/citation_signature","submit_replication":"https://pith.science/pith/FHE4BGGYIUITGQNNXURI7LPTKJ/action/replication_record"}},"created_at":"2026-07-05T11:57:07.997203+00:00","updated_at":"2026-07-05T11:57:07.997203+00:00"}