{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:4BA4ZXGQJWPT6DQ6YOUF3AKVLR","short_pith_number":"pith:4BA4ZXGQ","schema_version":"1.0","canonical_sha256":"e041ccdcd04d9f3f0e1ec3a85d81555c75cce004f500e9ab49a3cbbe0206b79c","source":{"kind":"arxiv","id":"2607.17715","version":1},"attestation_state":"computed","paper":{"title":"C$^2$KV: Compressed and Composable KV Cache Reuse for Efficient LLM Inference","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chaoyue Niu, Chuheng Du, Fan Wu, Guihai Chen, Hanlin Tang, Junyi Chen, Kan Liu, Lin Qu, Shengzhong Liu, Tao Lan","submitted_at":"2026-07-20T09:09:23Z","abstract_excerpt":"Long-context inference is central to modern large language model (LLM) applications such as retrieval-augmented generation and multi-document reasoning. To mitigate the growing inference cost, recent work has explored key-value (KV) cache reuse to reduce redundant prefill computation. However, existing reuse methods primarily focus on computation savings and overlook a critical bottleneck in long-context LLM serving: the cost of storing and accessing large KV caches. While KV compression appears to be a natural complement, naively combining compression with non-prefix KV reuse often leads to s"},"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":"2607.17715","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-20T09:09:23Z","cross_cats_sorted":[],"title_canon_sha256":"390614ddd3b128ec8676724970dc421b91a5d6f866d31069c5701d35d2b3ee2d","abstract_canon_sha256":"8f3237a47f0655cc9186aa526158a8c7442b37911f26234782adffc3585dcaca"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-21T02:21:56.019540Z","signature_b64":"ywKnU4s0gAdxSt1NIfKh/6AysUL/VtS3PLxdiNvt+x5Fzm0plu5mSWhIz4S6KKXyGLMQI1g695OPl/+oSDgbCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e041ccdcd04d9f3f0e1ec3a85d81555c75cce004f500e9ab49a3cbbe0206b79c","last_reissued_at":"2026-07-21T02:21:56.018386Z","signature_status":"signed_v1","first_computed_at":"2026-07-21T02:21:56.018386Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"C$^2$KV: Compressed and Composable KV Cache Reuse for Efficient LLM Inference","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chaoyue Niu, Chuheng Du, Fan Wu, Guihai Chen, Hanlin Tang, Junyi Chen, Kan Liu, Lin Qu, Shengzhong Liu, Tao Lan","submitted_at":"2026-07-20T09:09:23Z","abstract_excerpt":"Long-context inference is central to modern large language model (LLM) applications such as retrieval-augmented generation and multi-document reasoning. To mitigate the growing inference cost, recent work has explored key-value (KV) cache reuse to reduce redundant prefill computation. However, existing reuse methods primarily focus on computation savings and overlook a critical bottleneck in long-context LLM serving: the cost of storing and accessing large KV caches. While KV compression appears to be a natural complement, naively combining compression with non-prefix KV reuse often leads to s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.17715","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/2607.17715/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":"2607.17715","created_at":"2026-07-21T02:21:56.018838+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.17715v1","created_at":"2026-07-21T02:21:56.018838+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.17715","created_at":"2026-07-21T02:21:56.018838+00:00"},{"alias_kind":"pith_short_12","alias_value":"4BA4ZXGQJWPT","created_at":"2026-07-21T02:21:56.018838+00:00"},{"alias_kind":"pith_short_16","alias_value":"4BA4ZXGQJWPT6DQ6","created_at":"2026-07-21T02:21:56.018838+00:00"},{"alias_kind":"pith_short_8","alias_value":"4BA4ZXGQ","created_at":"2026-07-21T02:21:56.018838+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/4BA4ZXGQJWPT6DQ6YOUF3AKVLR","json":"https://pith.science/pith/4BA4ZXGQJWPT6DQ6YOUF3AKVLR.json","graph_json":"https://pith.science/api/pith-number/4BA4ZXGQJWPT6DQ6YOUF3AKVLR/graph.json","events_json":"https://pith.science/api/pith-number/4BA4ZXGQJWPT6DQ6YOUF3AKVLR/events.json","paper":"https://pith.science/paper/4BA4ZXGQ"},"agent_actions":{"view_html":"https://pith.science/pith/4BA4ZXGQJWPT6DQ6YOUF3AKVLR","download_json":"https://pith.science/pith/4BA4ZXGQJWPT6DQ6YOUF3AKVLR.json","view_paper":"https://pith.science/paper/4BA4ZXGQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.17715&json=true","fetch_graph":"https://pith.science/api/pith-number/4BA4ZXGQJWPT6DQ6YOUF3AKVLR/graph.json","fetch_events":"https://pith.science/api/pith-number/4BA4ZXGQJWPT6DQ6YOUF3AKVLR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4BA4ZXGQJWPT6DQ6YOUF3AKVLR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4BA4ZXGQJWPT6DQ6YOUF3AKVLR/action/storage_attestation","attest_author":"https://pith.science/pith/4BA4ZXGQJWPT6DQ6YOUF3AKVLR/action/author_attestation","sign_citation":"https://pith.science/pith/4BA4ZXGQJWPT6DQ6YOUF3AKVLR/action/citation_signature","submit_replication":"https://pith.science/pith/4BA4ZXGQJWPT6DQ6YOUF3AKVLR/action/replication_record"}},"created_at":"2026-07-21T02:21:56.018838+00:00","updated_at":"2026-07-21T02:21:56.018838+00:00"}