{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:YICLPSJF232SB7VKGB7U4MWTKK","short_pith_number":"pith:YICLPSJF","schema_version":"1.0","canonical_sha256":"c204b7c925d6f520feaa307f4e32d352ac02070f85511e7af900aa74579c56c0","source":{"kind":"arxiv","id":"2608.13263","version":1},"attestation_state":"computed","paper":{"title":"vToken: Token-Level Virtualization for Reclaimable KV Caches","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC","cs.OS"],"primary_cat":"cs.AI","authors_text":"Dongsheng Li, Hongjia Chen, Qianru Lv, Wenfei Wu, Xiangrui Yang, Yuanfeng Chen, Yuanhang Gao","submitted_at":"2026-08-13T14:01:21Z","abstract_excerpt":"Large language model serving faces a critical memory bottleneck: the KV cache grows with sequence length and batch size. PagedAttention uses fixed-size memory blocks to reduce allocator-level fragmentation, but recent KV eviction algorithms operate at a token granularity finer than block-level management. This mismatch causes intra-block fragmentation, leaving a large fraction of allocated KV memory unreclaimable. We present vToken, a lightweight token-level virtualization layer that decouples logical token liveness from physical block placement. vToken maintains a stable logical token view th"},"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":"2608.13263","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-08-13T14:01:21Z","cross_cats_sorted":["cs.DC","cs.OS"],"title_canon_sha256":"ae60da6a3943163df4698a17632902026304a4e540b4e0e3b57ace4ad176556b","abstract_canon_sha256":"acf7e2ec5d57e77c3b790a633e1556d7c3327fb956661834f2460f42c45a0698"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-14T01:00:48.506559Z","signature_b64":"0yqzUmxugHUx/kRe/PStuhNst5Qm0KoWK6PnbENjC7P8FIQFBkQTiv4vBC24gZiZTYSpu2qJLYeD/Pe/NDr8DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c204b7c925d6f520feaa307f4e32d352ac02070f85511e7af900aa74579c56c0","last_reissued_at":"2026-08-14T01:00:48.504484Z","signature_status":"signed_v1","first_computed_at":"2026-08-14T01:00:48.504484Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"vToken: Token-Level Virtualization for Reclaimable KV Caches","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC","cs.OS"],"primary_cat":"cs.AI","authors_text":"Dongsheng Li, Hongjia Chen, Qianru Lv, Wenfei Wu, Xiangrui Yang, Yuanfeng Chen, Yuanhang Gao","submitted_at":"2026-08-13T14:01:21Z","abstract_excerpt":"Large language model serving faces a critical memory bottleneck: the KV cache grows with sequence length and batch size. PagedAttention uses fixed-size memory blocks to reduce allocator-level fragmentation, but recent KV eviction algorithms operate at a token granularity finer than block-level management. This mismatch causes intra-block fragmentation, leaving a large fraction of allocated KV memory unreclaimable. We present vToken, a lightweight token-level virtualization layer that decouples logical token liveness from physical block placement. vToken maintains a stable logical token view th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.13263","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/2608.13263/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":"2608.13263","created_at":"2026-08-14T01:00:48.505531+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.13263v1","created_at":"2026-08-14T01:00:48.505531+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.13263","created_at":"2026-08-14T01:00:48.505531+00:00"},{"alias_kind":"pith_short_12","alias_value":"YICLPSJF232S","created_at":"2026-08-14T01:00:48.505531+00:00"},{"alias_kind":"pith_short_16","alias_value":"YICLPSJF232SB7VK","created_at":"2026-08-14T01:00:48.505531+00:00"},{"alias_kind":"pith_short_8","alias_value":"YICLPSJF","created_at":"2026-08-14T01:00:48.505531+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/YICLPSJF232SB7VKGB7U4MWTKK","json":"https://pith.science/pith/YICLPSJF232SB7VKGB7U4MWTKK.json","graph_json":"https://pith.science/api/pith-number/YICLPSJF232SB7VKGB7U4MWTKK/graph.json","events_json":"https://pith.science/api/pith-number/YICLPSJF232SB7VKGB7U4MWTKK/events.json","paper":"https://pith.science/paper/YICLPSJF"},"agent_actions":{"view_html":"https://pith.science/pith/YICLPSJF232SB7VKGB7U4MWTKK","download_json":"https://pith.science/pith/YICLPSJF232SB7VKGB7U4MWTKK.json","view_paper":"https://pith.science/paper/YICLPSJF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.13263&json=true","fetch_graph":"https://pith.science/api/pith-number/YICLPSJF232SB7VKGB7U4MWTKK/graph.json","fetch_events":"https://pith.science/api/pith-number/YICLPSJF232SB7VKGB7U4MWTKK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YICLPSJF232SB7VKGB7U4MWTKK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YICLPSJF232SB7VKGB7U4MWTKK/action/storage_attestation","attest_author":"https://pith.science/pith/YICLPSJF232SB7VKGB7U4MWTKK/action/author_attestation","sign_citation":"https://pith.science/pith/YICLPSJF232SB7VKGB7U4MWTKK/action/citation_signature","submit_replication":"https://pith.science/pith/YICLPSJF232SB7VKGB7U4MWTKK/action/replication_record"}},"created_at":"2026-08-14T01:00:48.505531+00:00","updated_at":"2026-08-14T01:00:48.505531+00:00"}