{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:XZSBIJGQMWOOHGSQBRHWQV65IG","short_pith_number":"pith:XZSBIJGQ","canonical_record":{"source":{"id":"2311.04934","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-11-07T18:17:05Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9f3977bf5b278c6926d72643eace933bd7adba2afda577452fb7d892be7c8b4b","abstract_canon_sha256":"e560f0a88acc8c91f73090d22acc42e24b7114b519bc799fa960323edd94b50a"},"schema_version":"1.0"},"canonical_sha256":"be641424d0659ce39a500c4f6857dd41a59d7fe0c4e5cc58e7a99b64c6121890","source":{"kind":"arxiv","id":"2311.04934","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2311.04934","created_at":"2026-07-05T08:11:55Z"},{"alias_kind":"arxiv_version","alias_value":"2311.04934v2","created_at":"2026-07-05T08:11:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.04934","created_at":"2026-07-05T08:11:55Z"},{"alias_kind":"pith_short_12","alias_value":"XZSBIJGQMWOO","created_at":"2026-07-05T08:11:55Z"},{"alias_kind":"pith_short_16","alias_value":"XZSBIJGQMWOOHGSQ","created_at":"2026-07-05T08:11:55Z"},{"alias_kind":"pith_short_8","alias_value":"XZSBIJGQ","created_at":"2026-07-05T08:11:55Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:XZSBIJGQMWOOHGSQBRHWQV65IG","target":"record","payload":{"canonical_record":{"source":{"id":"2311.04934","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-11-07T18:17:05Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9f3977bf5b278c6926d72643eace933bd7adba2afda577452fb7d892be7c8b4b","abstract_canon_sha256":"e560f0a88acc8c91f73090d22acc42e24b7114b519bc799fa960323edd94b50a"},"schema_version":"1.0"},"canonical_sha256":"be641424d0659ce39a500c4f6857dd41a59d7fe0c4e5cc58e7a99b64c6121890","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:11:55.376153Z","signature_b64":"xH9yTITGV00mLJX9NL1dSgwpu/6n8yfSkVuDOkxbQ4wnSHAHnRAUXYyMY2LeVa5GZGgssAYTdB2YvQOQ3jigBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"be641424d0659ce39a500c4f6857dd41a59d7fe0c4e5cc58e7a99b64c6121890","last_reissued_at":"2026-07-05T08:11:55.375709Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:11:55.375709Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2311.04934","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:11:55Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ElC29QEHe1v+AqtgR/cRCu24gooseputcTN2LAvvQSiYzjZUDN4pGQts0HLjgR3xfo2CVAsy/CPb3i5S9Tk3Dg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T14:14:28.725486Z"},"content_sha256":"8d29a6454f43240722c4dc1c49176fb6a031ae749e0cfb2a6da7b37765195b85","schema_version":"1.0","event_id":"sha256:8d29a6454f43240722c4dc1c49176fb6a031ae749e0cfb2a6da7b37765195b85"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:XZSBIJGQMWOOHGSQBRHWQV65IG","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Prompt Cache: Modular Attention Reuse for Low-Latency Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Anurag Khandelwal, Guojun Chen, In Gim, Lin Zhong, Nikhil Sarda, Seung-seob Lee","submitted_at":"2023-11-07T18:17:05Z","abstract_excerpt":"We present Prompt Cache, an approach for accelerating inference for large language models (LLM) by reusing attention states across different LLM prompts. Many input prompts have overlapping text segments, such as system messages, prompt templates, and documents provided for context. Our key insight is that by precomputing and storing the attention states of these frequently occurring text segments on the inference server, we can efficiently reuse them when these segments appear in user prompts. Prompt Cache employs a schema to explicitly define such reusable text segments, called prompt module"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.04934","kind":"arxiv","version":2},"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/2311.04934/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:11:55Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Yot01V9xS2ORQ0oh36znFAe0EbuqSvyX0d0IQqidQZ0tvx/VIs+Rl2sW39HqTQp2tN5f+1DZRb55yc7SH1sqDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T14:14:28.726328Z"},"content_sha256":"52dcf0da51010cd463e784b9b78a63862044cd4d67738d4ecd4a5bd834b562fa","schema_version":"1.0","event_id":"sha256:52dcf0da51010cd463e784b9b78a63862044cd4d67738d4ecd4a5bd834b562fa"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/XZSBIJGQMWOOHGSQBRHWQV65IG/bundle.json","state_url":"https://pith.science/pith/XZSBIJGQMWOOHGSQBRHWQV65IG/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/XZSBIJGQMWOOHGSQBRHWQV65IG/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-17T14:14:28Z","links":{"resolver":"https://pith.science/pith/XZSBIJGQMWOOHGSQBRHWQV65IG","bundle":"https://pith.science/pith/XZSBIJGQMWOOHGSQBRHWQV65IG/bundle.json","state":"https://pith.science/pith/XZSBIJGQMWOOHGSQBRHWQV65IG/state.json","well_known_bundle":"https://pith.science/.well-known/pith/XZSBIJGQMWOOHGSQBRHWQV65IG/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:XZSBIJGQMWOOHGSQBRHWQV65IG","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":"e560f0a88acc8c91f73090d22acc42e24b7114b519bc799fa960323edd94b50a","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-11-07T18:17:05Z","title_canon_sha256":"9f3977bf5b278c6926d72643eace933bd7adba2afda577452fb7d892be7c8b4b"},"schema_version":"1.0","source":{"id":"2311.04934","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2311.04934","created_at":"2026-07-05T08:11:55Z"},{"alias_kind":"arxiv_version","alias_value":"2311.04934v2","created_at":"2026-07-05T08:11:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.04934","created_at":"2026-07-05T08:11:55Z"},{"alias_kind":"pith_short_12","alias_value":"XZSBIJGQMWOO","created_at":"2026-07-05T08:11:55Z"},{"alias_kind":"pith_short_16","alias_value":"XZSBIJGQMWOOHGSQ","created_at":"2026-07-05T08:11:55Z"},{"alias_kind":"pith_short_8","alias_value":"XZSBIJGQ","created_at":"2026-07-05T08:11:55Z"}],"graph_snapshots":[{"event_id":"sha256:52dcf0da51010cd463e784b9b78a63862044cd4d67738d4ecd4a5bd834b562fa","target":"graph","created_at":"2026-07-05T08:11:55Z","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/2311.04934/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We present Prompt Cache, an approach for accelerating inference for large language models (LLM) by reusing attention states across different LLM prompts. Many input prompts have overlapping text segments, such as system messages, prompt templates, and documents provided for context. Our key insight is that by precomputing and storing the attention states of these frequently occurring text segments on the inference server, we can efficiently reuse them when these segments appear in user prompts. Prompt Cache employs a schema to explicitly define such reusable text segments, called prompt module","authors_text":"Anurag Khandelwal, Guojun Chen, In Gim, Lin Zhong, Nikhil Sarda, Seung-seob Lee","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-11-07T18:17:05Z","title":"Prompt Cache: Modular Attention Reuse for Low-Latency Inference"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.04934","kind":"arxiv","version":2},"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":"sha256:8d29a6454f43240722c4dc1c49176fb6a031ae749e0cfb2a6da7b37765195b85","target":"record","created_at":"2026-07-05T08:11:55Z","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":"e560f0a88acc8c91f73090d22acc42e24b7114b519bc799fa960323edd94b50a","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-11-07T18:17:05Z","title_canon_sha256":"9f3977bf5b278c6926d72643eace933bd7adba2afda577452fb7d892be7c8b4b"},"schema_version":"1.0","source":{"id":"2311.04934","kind":"arxiv","version":2}},"canonical_sha256":"be641424d0659ce39a500c4f6857dd41a59d7fe0c4e5cc58e7a99b64c6121890","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"be641424d0659ce39a500c4f6857dd41a59d7fe0c4e5cc58e7a99b64c6121890","first_computed_at":"2026-07-05T08:11:55.375709Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:11:55.375709Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"xH9yTITGV00mLJX9NL1dSgwpu/6n8yfSkVuDOkxbQ4wnSHAHnRAUXYyMY2LeVa5GZGgssAYTdB2YvQOQ3jigBA==","signature_status":"signed_v1","signed_at":"2026-07-05T08:11:55.376153Z","signed_message":"canonical_sha256_bytes"},"source_id":"2311.04934","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8d29a6454f43240722c4dc1c49176fb6a031ae749e0cfb2a6da7b37765195b85","sha256:52dcf0da51010cd463e784b9b78a63862044cd4d67738d4ecd4a5bd834b562fa"],"state_sha256":"c51f5f447dd9aa0718169d56a57bd099e73196143554d004c7b67a5d12718a12"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5NG1w3nFGpsMgHIrY75exa/Eh3qQvKlE/Drr9Zo7x6NKiN5GsTgYmV3GtFllDt0yPGnC5Cv5KqhGQARDghRrBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-17T14:14:28.732611Z","bundle_sha256":"4d31c82ba861ffc749051ddb3cb4e7137d9b33dcb6b38d04209ac110ee0e74af"}}