{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:53GA4MRJIWJ7M3LUGMMZCB674H","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"0079e4bd3e1525e45ed01fece839223fd4497f45aae98e2362a2f8517ea8f433","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-22T07:52:38Z","title_canon_sha256":"6a23985ce2b346dc48f8805b2ad92513a986881b975df2d79324e9cc751569e7"},"schema_version":"1.0","source":{"id":"2501.12689","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.12689","created_at":"2026-07-05T12:04:30Z"},{"alias_kind":"arxiv_version","alias_value":"2501.12689v3","created_at":"2026-07-05T12:04:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.12689","created_at":"2026-07-05T12:04:30Z"},{"alias_kind":"pith_short_12","alias_value":"53GA4MRJIWJ7","created_at":"2026-07-05T12:04:30Z"},{"alias_kind":"pith_short_16","alias_value":"53GA4MRJIWJ7M3LU","created_at":"2026-07-05T12:04:30Z"},{"alias_kind":"pith_short_8","alias_value":"53GA4MRJ","created_at":"2026-07-05T12:04:30Z"}],"graph_snapshots":[{"event_id":"sha256:075449990a552ca10240a0dc63f0f92da4e37ff52fdf55181cbef102f6058b71","target":"graph","created_at":"2026-07-05T12:04:30Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2501.12689/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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","authors_text":"Arvind Krishnamurthy, David Culler, Fan Lai, Henry M. Levy, Jiaming Shen, Lillian Tsai, Nikhil Sarda, Yanqi Zhou, Yifan Yu, Yu Gan","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-22T07:52:38Z","title":"IC-Cache: Efficient Large Language Model Serving via In-context Caching"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.12689","kind":"arxiv","version":3},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:452ce7867c24e6e514dcb8420c3f540881d7d19ceaf16e335df7351cc135273e","target":"record","created_at":"2026-07-05T12:04:30Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"0079e4bd3e1525e45ed01fece839223fd4497f45aae98e2362a2f8517ea8f433","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-22T07:52:38Z","title_canon_sha256":"6a23985ce2b346dc48f8805b2ad92513a986881b975df2d79324e9cc751569e7"},"schema_version":"1.0","source":{"id":"2501.12689","kind":"arxiv","version":3}},"canonical_sha256":"eecc0e32294593f66d7433199107dfe1db5d206b5e342384d7f0378e42a7cd08","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"eecc0e32294593f66d7433199107dfe1db5d206b5e342384d7f0378e42a7cd08","first_computed_at":"2026-07-05T12:04:30.415082Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T12:04:30.415082Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"sL8k+C6lU1+0vEgxbnVonGwd3/uDG6V8+dZO48/DAp09Bj3IwjkGxnQYI2Xncb0b2hE4UrxIEj84z4EAdv5mCg==","signature_status":"signed_v1","signed_at":"2026-07-05T12:04:30.415584Z","signed_message":"canonical_sha256_bytes"},"source_id":"2501.12689","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:452ce7867c24e6e514dcb8420c3f540881d7d19ceaf16e335df7351cc135273e","sha256:075449990a552ca10240a0dc63f0f92da4e37ff52fdf55181cbef102f6058b71"],"state_sha256":"fec9582d6d710f3ebb58fcbf391b71eb36a63df46402f9473ece740474c87d32"}