{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:FPDXDU6IMGOXCUSNYA5SR2I5NE","short_pith_number":"pith:FPDXDU6I","schema_version":"1.0","canonical_sha256":"2bc771d3c8619d71524dc03b28e91d6900f70f1704c06a6a563c18a25c1d93ff","source":{"kind":"arxiv","id":"2606.05875","version":1},"attestation_state":"computed","paper":{"title":"QCFuse: Query-Aware Cache Fusion via Compressed View for Efficient RAG Serving","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DB"],"primary_cat":"cs.AI","authors_text":"Haoyang Li, Jiabao Jin, Jianxin Yan, Jia Zhu, Kui Ren, Lei Chen, Peng Cheng, Wangze Ni, Xuemin Lin, Zhenxin Li, Zhitao Shen","submitted_at":"2026-06-04T08:47:46Z","abstract_excerpt":"Retrieval-augmented generation (RAG) improves large language model (LLM) answer quality by grounding generation in external evidence, but processing retrieved contexts makes the prefill stage a dominant serving cost. RAG cache fusion reduces this cost by reusing precomputed key-value (KV) caches for retrieved chunks and selectively recomputing tokens under the current prompt. Existing selectors, however, face a dilemma between quality and efficiency: fast query-agnostic or final-layer query-to-context selectors can miss request-relevant evidence, whereas full-view query-aware selectors require"},"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":"2606.05875","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-04T08:47:46Z","cross_cats_sorted":["cs.DB"],"title_canon_sha256":"ef1d33bc7dfdb852b39ac9b8aab808da08791d3a9db62786d32f5c1adaf19c8d","abstract_canon_sha256":"6a9596d9d791f87e22fcc774e3b2d74a45a7f8f82d3bf4c0ff3b6cc4e9ffcf80"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-05T01:15:06.532482Z","signature_b64":"Qevnxp0hnQ+ZHwWpFdNjmJfBU+yv/ywpfBXloxbg3rMR9eGy6GZY0KLN24gdcdtQsQLeRNGzu+hckvTM8oJ4CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2bc771d3c8619d71524dc03b28e91d6900f70f1704c06a6a563c18a25c1d93ff","last_reissued_at":"2026-06-05T01:15:06.532083Z","signature_status":"signed_v1","first_computed_at":"2026-06-05T01:15:06.532083Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"QCFuse: Query-Aware Cache Fusion via Compressed View for Efficient RAG Serving","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DB"],"primary_cat":"cs.AI","authors_text":"Haoyang Li, Jiabao Jin, Jianxin Yan, Jia Zhu, Kui Ren, Lei Chen, Peng Cheng, Wangze Ni, Xuemin Lin, Zhenxin Li, Zhitao Shen","submitted_at":"2026-06-04T08:47:46Z","abstract_excerpt":"Retrieval-augmented generation (RAG) improves large language model (LLM) answer quality by grounding generation in external evidence, but processing retrieved contexts makes the prefill stage a dominant serving cost. RAG cache fusion reduces this cost by reusing precomputed key-value (KV) caches for retrieved chunks and selectively recomputing tokens under the current prompt. Existing selectors, however, face a dilemma between quality and efficiency: fast query-agnostic or final-layer query-to-context selectors can miss request-relevant evidence, whereas full-view query-aware selectors require"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.05875","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/2606.05875/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":"2606.05875","created_at":"2026-06-05T01:15:06.532139+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.05875v1","created_at":"2026-06-05T01:15:06.532139+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.05875","created_at":"2026-06-05T01:15:06.532139+00:00"},{"alias_kind":"pith_short_12","alias_value":"FPDXDU6IMGOX","created_at":"2026-06-05T01:15:06.532139+00:00"},{"alias_kind":"pith_short_16","alias_value":"FPDXDU6IMGOXCUSN","created_at":"2026-06-05T01:15:06.532139+00:00"},{"alias_kind":"pith_short_8","alias_value":"FPDXDU6I","created_at":"2026-06-05T01:15:06.532139+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/FPDXDU6IMGOXCUSNYA5SR2I5NE","json":"https://pith.science/pith/FPDXDU6IMGOXCUSNYA5SR2I5NE.json","graph_json":"https://pith.science/api/pith-number/FPDXDU6IMGOXCUSNYA5SR2I5NE/graph.json","events_json":"https://pith.science/api/pith-number/FPDXDU6IMGOXCUSNYA5SR2I5NE/events.json","paper":"https://pith.science/paper/FPDXDU6I"},"agent_actions":{"view_html":"https://pith.science/pith/FPDXDU6IMGOXCUSNYA5SR2I5NE","download_json":"https://pith.science/pith/FPDXDU6IMGOXCUSNYA5SR2I5NE.json","view_paper":"https://pith.science/paper/FPDXDU6I","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.05875&json=true","fetch_graph":"https://pith.science/api/pith-number/FPDXDU6IMGOXCUSNYA5SR2I5NE/graph.json","fetch_events":"https://pith.science/api/pith-number/FPDXDU6IMGOXCUSNYA5SR2I5NE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FPDXDU6IMGOXCUSNYA5SR2I5NE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FPDXDU6IMGOXCUSNYA5SR2I5NE/action/storage_attestation","attest_author":"https://pith.science/pith/FPDXDU6IMGOXCUSNYA5SR2I5NE/action/author_attestation","sign_citation":"https://pith.science/pith/FPDXDU6IMGOXCUSNYA5SR2I5NE/action/citation_signature","submit_replication":"https://pith.science/pith/FPDXDU6IMGOXCUSNYA5SR2I5NE/action/replication_record"}},"created_at":"2026-06-05T01:15:06.532139+00:00","updated_at":"2026-06-05T01:15:06.532139+00:00"}