{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:EUTHUZPZPUU3VKIXTMURTZFJ5E","short_pith_number":"pith:EUTHUZPZ","schema_version":"1.0","canonical_sha256":"25267a65f97d29baa9179b2919e4a9e90a70a5c15f7af917874831a820df2259","source":{"kind":"arxiv","id":"2608.01526","version":1},"attestation_state":"computed","paper":{"title":"An Internet for the KV Cache: Rethinking Classical Infrastructure Boundaries in the LLM Inference Age","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.NI","authors_text":"Junchen Jiang, Nick Feamster, Siddhant Ray","submitted_at":"2026-08-02T22:31:22Z","abstract_excerpt":"LLM inference has become a global-scale, heterogeneous workload spanning agents, retrieval, tool-use, code execution and multi-modal reasoning. These workloads naturally enable context reuse from overlapping inputs, creating a major opportunity to store and reuse the contexts' KV Caches instead of recomputing them. However, model-side advances that shrink the KV Cache and system-side advances that reduce compute, storage, and transfer costs are evolve independently within legacy cloud boundaries. We argue that future inference infrastructure should allow decoupling of compute and KV Cache stor"},"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":"2608.01526","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.NI","submitted_at":"2026-08-02T22:31:22Z","cross_cats_sorted":[],"title_canon_sha256":"ed02688437a1dee65d21170def56e82e3aa43989564e9e029bbfbe2dbd1bc2fb","abstract_canon_sha256":"f0ee9579f8f4c6e22ad03682d0e00d97118762a4e08e3fc39050704ada23088e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-04T02:05:48.472790Z","signature_b64":"3cJnEtFE0gNRyBZ/0E7NmTIL5bfT+8J8l9Jc71f1VsWxDq3z59O9GEGAww3Gumg4CzIr5LbBBPIJ5kNuIE50AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"25267a65f97d29baa9179b2919e4a9e90a70a5c15f7af917874831a820df2259","last_reissued_at":"2026-08-04T02:05:48.470770Z","signature_status":"signed_v1","first_computed_at":"2026-08-04T02:05:48.470770Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"An Internet for the KV Cache: Rethinking Classical Infrastructure Boundaries in the LLM Inference Age","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.NI","authors_text":"Junchen Jiang, Nick Feamster, Siddhant Ray","submitted_at":"2026-08-02T22:31:22Z","abstract_excerpt":"LLM inference has become a global-scale, heterogeneous workload spanning agents, retrieval, tool-use, code execution and multi-modal reasoning. These workloads naturally enable context reuse from overlapping inputs, creating a major opportunity to store and reuse the contexts' KV Caches instead of recomputing them. However, model-side advances that shrink the KV Cache and system-side advances that reduce compute, storage, and transfer costs are evolve independently within legacy cloud boundaries. We argue that future inference infrastructure should allow decoupling of compute and KV Cache stor"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.01526","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/2608.01526/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":"2608.01526","created_at":"2026-08-04T02:05:48.472409+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.01526v1","created_at":"2026-08-04T02:05:48.472409+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.01526","created_at":"2026-08-04T02:05:48.472409+00:00"},{"alias_kind":"pith_short_12","alias_value":"EUTHUZPZPUU3","created_at":"2026-08-04T02:05:48.472409+00:00"},{"alias_kind":"pith_short_16","alias_value":"EUTHUZPZPUU3VKIX","created_at":"2026-08-04T02:05:48.472409+00:00"},{"alias_kind":"pith_short_8","alias_value":"EUTHUZPZ","created_at":"2026-08-04T02:05:48.472409+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/EUTHUZPZPUU3VKIXTMURTZFJ5E","json":"https://pith.science/pith/EUTHUZPZPUU3VKIXTMURTZFJ5E.json","graph_json":"https://pith.science/api/pith-number/EUTHUZPZPUU3VKIXTMURTZFJ5E/graph.json","events_json":"https://pith.science/api/pith-number/EUTHUZPZPUU3VKIXTMURTZFJ5E/events.json","paper":"https://pith.science/paper/EUTHUZPZ"},"agent_actions":{"view_html":"https://pith.science/pith/EUTHUZPZPUU3VKIXTMURTZFJ5E","download_json":"https://pith.science/pith/EUTHUZPZPUU3VKIXTMURTZFJ5E.json","view_paper":"https://pith.science/paper/EUTHUZPZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.01526&json=true","fetch_graph":"https://pith.science/api/pith-number/EUTHUZPZPUU3VKIXTMURTZFJ5E/graph.json","fetch_events":"https://pith.science/api/pith-number/EUTHUZPZPUU3VKIXTMURTZFJ5E/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EUTHUZPZPUU3VKIXTMURTZFJ5E/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EUTHUZPZPUU3VKIXTMURTZFJ5E/action/storage_attestation","attest_author":"https://pith.science/pith/EUTHUZPZPUU3VKIXTMURTZFJ5E/action/author_attestation","sign_citation":"https://pith.science/pith/EUTHUZPZPUU3VKIXTMURTZFJ5E/action/citation_signature","submit_replication":"https://pith.science/pith/EUTHUZPZPUU3VKIXTMURTZFJ5E/action/replication_record"}},"created_at":"2026-08-04T02:05:48.472409+00:00","updated_at":"2026-08-04T02:05:48.472409+00:00"}