{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:53GA4MRJIWJ7M3LUGMMZCB674H","short_pith_number":"pith:53GA4MRJ","schema_version":"1.0","canonical_sha256":"eecc0e32294593f66d7433199107dfe1db5d206b5e342384d7f0378e42a7cd08","source":{"kind":"arxiv","id":"2501.12689","version":3},"attestation_state":"computed","paper":{"title":"IC-Cache: Efficient Large Language Model Serving via In-context Caching","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Arvind Krishnamurthy, David Culler, Fan Lai, Henry M. Levy, Jiaming Shen, Lillian Tsai, Nikhil Sarda, Yanqi Zhou, Yifan Yu, Yu Gan","submitted_at":"2025-01-22T07:52:38Z","abstract_excerpt":"Large language models (LLMs) have excelled in various applications, yet serving them at scale is challenging due to their substantial resource demands and high latency. Our real-world studies reveal that over 70% of user requests to LLMs have semantically similar counterparts, suggesting the potential for knowledge transfer among requests. However, naively caching and reusing past responses leads to a big quality drop. In this paper, we introduce IC-Cache, a caching system that enables live LLM capability augmentation to improve serving efficiency: by leveraging historical request-response pai"},"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":"2501.12689","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-22T07:52:38Z","cross_cats_sorted":[],"title_canon_sha256":"6a23985ce2b346dc48f8805b2ad92513a986881b975df2d79324e9cc751569e7","abstract_canon_sha256":"0079e4bd3e1525e45ed01fece839223fd4497f45aae98e2362a2f8517ea8f433"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:04:30.415584Z","signature_b64":"sL8k+C6lU1+0vEgxbnVonGwd3/uDG6V8+dZO48/DAp09Bj3IwjkGxnQYI2Xncb0b2hE4UrxIEj84z4EAdv5mCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"eecc0e32294593f66d7433199107dfe1db5d206b5e342384d7f0378e42a7cd08","last_reissued_at":"2026-07-05T12:04:30.415082Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:04:30.415082Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"IC-Cache: Efficient Large Language Model Serving via In-context Caching","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Arvind Krishnamurthy, David Culler, Fan Lai, Henry M. Levy, Jiaming Shen, Lillian Tsai, Nikhil Sarda, Yanqi Zhou, Yifan Yu, Yu Gan","submitted_at":"2025-01-22T07:52:38Z","abstract_excerpt":"Large language models (LLMs) have excelled in various applications, yet serving them at scale is challenging due to their substantial resource demands and high latency. Our real-world studies reveal that over 70% of user requests to LLMs have semantically similar counterparts, suggesting the potential for knowledge transfer among requests. However, naively caching and reusing past responses leads to a big quality drop. In this paper, we introduce IC-Cache, a caching system that enables live LLM capability augmentation to improve serving efficiency: by leveraging historical request-response pai"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.12689","kind":"arxiv","version":3},"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/2501.12689/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":"2501.12689","created_at":"2026-07-05T12:04:30.415142+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.12689v3","created_at":"2026-07-05T12:04:30.415142+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.12689","created_at":"2026-07-05T12:04:30.415142+00:00"},{"alias_kind":"pith_short_12","alias_value":"53GA4MRJIWJ7","created_at":"2026-07-05T12:04:30.415142+00:00"},{"alias_kind":"pith_short_16","alias_value":"53GA4MRJIWJ7M3LU","created_at":"2026-07-05T12:04:30.415142+00:00"},{"alias_kind":"pith_short_8","alias_value":"53GA4MRJ","created_at":"2026-07-05T12:04:30.415142+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/53GA4MRJIWJ7M3LUGMMZCB674H","json":"https://pith.science/pith/53GA4MRJIWJ7M3LUGMMZCB674H.json","graph_json":"https://pith.science/api/pith-number/53GA4MRJIWJ7M3LUGMMZCB674H/graph.json","events_json":"https://pith.science/api/pith-number/53GA4MRJIWJ7M3LUGMMZCB674H/events.json","paper":"https://pith.science/paper/53GA4MRJ"},"agent_actions":{"view_html":"https://pith.science/pith/53GA4MRJIWJ7M3LUGMMZCB674H","download_json":"https://pith.science/pith/53GA4MRJIWJ7M3LUGMMZCB674H.json","view_paper":"https://pith.science/paper/53GA4MRJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.12689&json=true","fetch_graph":"https://pith.science/api/pith-number/53GA4MRJIWJ7M3LUGMMZCB674H/graph.json","fetch_events":"https://pith.science/api/pith-number/53GA4MRJIWJ7M3LUGMMZCB674H/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/53GA4MRJIWJ7M3LUGMMZCB674H/action/timestamp_anchor","attest_storage":"https://pith.science/pith/53GA4MRJIWJ7M3LUGMMZCB674H/action/storage_attestation","attest_author":"https://pith.science/pith/53GA4MRJIWJ7M3LUGMMZCB674H/action/author_attestation","sign_citation":"https://pith.science/pith/53GA4MRJIWJ7M3LUGMMZCB674H/action/citation_signature","submit_replication":"https://pith.science/pith/53GA4MRJIWJ7M3LUGMMZCB674H/action/replication_record"}},"created_at":"2026-07-05T12:04:30.415142+00:00","updated_at":"2026-07-05T12:04:30.415142+00:00"}