{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:TBPNRYTTUM2MTA2BHN2C2YD7PH","short_pith_number":"pith:TBPNRYTT","schema_version":"1.0","canonical_sha256":"985ed8e273a334c983413b742d607f79f6681b64ff262171e92b243bd5d883af","source":{"kind":"arxiv","id":"2608.02376","version":1},"attestation_state":"computed","paper":{"title":"Token-Native Storage: Read and Write in your Agent's Language","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.IR"],"primary_cat":"cs.DB","authors_text":"Kumar Shivendu","submitted_at":"2026-08-03T15:20:36Z","abstract_excerpt":"Search and database engines still store text as UTF-8, a format built for humans. But the systems that increasingly read and write that text (embedders, rerankers, and language-model agents) work in token IDs, not characters, so every access pays to translate between the two. As agents become the primary readers and writers of stored text, we argue for token-native storage: keep the text as the model's own byte-pair-encoding (BPE) token IDs. This is both smaller and faster. Packing r50k IDs as uint16 already beats UTF-8 by 2.25x on English with no compression, and an entropy coder reaches 3.30"},"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":"2608.02376","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DB","submitted_at":"2026-08-03T15:20:36Z","cross_cats_sorted":["cs.CL","cs.IR"],"title_canon_sha256":"3bb550ea08ee7bdec2ada379349a782bc0891886349d38eb42a03e2ffe0b97bd","abstract_canon_sha256":"41ef218fbe167c7118af4b08626a91f9513ac670c36b84a86743afb2948f8c0d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-04T02:12:05.771487Z","signature_b64":"8wx+Ujgc3u4OQw3sa5XnYTvuMlbTTFErb6bqVyF9S3k7uE0dCW0DLaaWiNfFSydswuHHLEiNNjuWI1DEoEzcCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"985ed8e273a334c983413b742d607f79f6681b64ff262171e92b243bd5d883af","last_reissued_at":"2026-08-04T02:12:05.769826Z","signature_status":"signed_v1","first_computed_at":"2026-08-04T02:12:05.769826Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Token-Native Storage: Read and Write in your Agent's Language","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.IR"],"primary_cat":"cs.DB","authors_text":"Kumar Shivendu","submitted_at":"2026-08-03T15:20:36Z","abstract_excerpt":"Search and database engines still store text as UTF-8, a format built for humans. But the systems that increasingly read and write that text (embedders, rerankers, and language-model agents) work in token IDs, not characters, so every access pays to translate between the two. As agents become the primary readers and writers of stored text, we argue for token-native storage: keep the text as the model's own byte-pair-encoding (BPE) token IDs. This is both smaller and faster. Packing r50k IDs as uint16 already beats UTF-8 by 2.25x on English with no compression, and an entropy coder reaches 3.30"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.02376","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/2608.02376/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":"2608.02376","created_at":"2026-08-04T02:12:05.771389+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.02376v1","created_at":"2026-08-04T02:12:05.771389+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.02376","created_at":"2026-08-04T02:12:05.771389+00:00"},{"alias_kind":"pith_short_12","alias_value":"TBPNRYTTUM2M","created_at":"2026-08-04T02:12:05.771389+00:00"},{"alias_kind":"pith_short_16","alias_value":"TBPNRYTTUM2MTA2B","created_at":"2026-08-04T02:12:05.771389+00:00"},{"alias_kind":"pith_short_8","alias_value":"TBPNRYTT","created_at":"2026-08-04T02:12:05.771389+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/TBPNRYTTUM2MTA2BHN2C2YD7PH","json":"https://pith.science/pith/TBPNRYTTUM2MTA2BHN2C2YD7PH.json","graph_json":"https://pith.science/api/pith-number/TBPNRYTTUM2MTA2BHN2C2YD7PH/graph.json","events_json":"https://pith.science/api/pith-number/TBPNRYTTUM2MTA2BHN2C2YD7PH/events.json","paper":"https://pith.science/paper/TBPNRYTT"},"agent_actions":{"view_html":"https://pith.science/pith/TBPNRYTTUM2MTA2BHN2C2YD7PH","download_json":"https://pith.science/pith/TBPNRYTTUM2MTA2BHN2C2YD7PH.json","view_paper":"https://pith.science/paper/TBPNRYTT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.02376&json=true","fetch_graph":"https://pith.science/api/pith-number/TBPNRYTTUM2MTA2BHN2C2YD7PH/graph.json","fetch_events":"https://pith.science/api/pith-number/TBPNRYTTUM2MTA2BHN2C2YD7PH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TBPNRYTTUM2MTA2BHN2C2YD7PH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TBPNRYTTUM2MTA2BHN2C2YD7PH/action/storage_attestation","attest_author":"https://pith.science/pith/TBPNRYTTUM2MTA2BHN2C2YD7PH/action/author_attestation","sign_citation":"https://pith.science/pith/TBPNRYTTUM2MTA2BHN2C2YD7PH/action/citation_signature","submit_replication":"https://pith.science/pith/TBPNRYTTUM2MTA2BHN2C2YD7PH/action/replication_record"}},"created_at":"2026-08-04T02:12:05.771389+00:00","updated_at":"2026-08-04T02:12:05.771389+00:00"}