{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:IDJAH5RTFLKMAA4YIROEVE6D5T","short_pith_number":"pith:IDJAH5RT","canonical_record":{"source":{"id":"2405.05329","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.DC","submitted_at":"2024-05-08T18:03:22Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"8d5ac19dc47aed5ea37529ce7eba8339073cf99f68a891cf2c451eb388d8ed42","abstract_canon_sha256":"b0ba9ea19707a44dbc6651c023e3a7d0d584e66deb33d77c8e110c5a61336bdd"},"schema_version":"1.0"},"canonical_sha256":"40d203f6332ad4c00398445c4a93c3ecd75df50a3ae8e242a096e038ede6d596","source":{"kind":"arxiv","id":"2405.05329","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.05329","created_at":"2026-07-05T08:18:52Z"},{"alias_kind":"arxiv_version","alias_value":"2405.05329v2","created_at":"2026-07-05T08:18:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.05329","created_at":"2026-07-05T08:18:52Z"},{"alias_kind":"pith_short_12","alias_value":"IDJAH5RTFLKM","created_at":"2026-07-05T08:18:52Z"},{"alias_kind":"pith_short_16","alias_value":"IDJAH5RTFLKMAA4Y","created_at":"2026-07-05T08:18:52Z"},{"alias_kind":"pith_short_8","alias_value":"IDJAH5RT","created_at":"2026-07-05T08:18:52Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:IDJAH5RTFLKMAA4YIROEVE6D5T","target":"record","payload":{"canonical_record":{"source":{"id":"2405.05329","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.DC","submitted_at":"2024-05-08T18:03:22Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"8d5ac19dc47aed5ea37529ce7eba8339073cf99f68a891cf2c451eb388d8ed42","abstract_canon_sha256":"b0ba9ea19707a44dbc6651c023e3a7d0d584e66deb33d77c8e110c5a61336bdd"},"schema_version":"1.0"},"canonical_sha256":"40d203f6332ad4c00398445c4a93c3ecd75df50a3ae8e242a096e038ede6d596","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:18:52.754625Z","signature_b64":"SizCOM9dO+1F3xh4dhu2jYnIAXswMjQ5V04ZhPFTH1bVptt9t4Nq1NGcXqfmx6yFP22tECBYVKdAlm0RvAaaDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"40d203f6332ad4c00398445c4a93c3ecd75df50a3ae8e242a096e038ede6d596","last_reissued_at":"2026-07-05T08:18:52.754188Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:18:52.754188Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2405.05329","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":"2026-07-05T08:18:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IPWiFGjtKLJCFglVGj7zE+tS9QIPJLgCgN7d/Ak//Hxj86wFI6tF5RNT6toM/g6AnO4BFazI5XTUctdRl/+bBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T14:49:46.623114Z"},"content_sha256":"7d17bf4afba3f4d14ff2dd3c68c20b8c3c3878ee8fdc97d7a59329a8dbfaa4a3","schema_version":"1.0","event_id":"sha256:7d17bf4afba3f4d14ff2dd3c68c20b8c3c3878ee8fdc97d7a59329a8dbfaa4a3"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:IDJAH5RTFLKMAA4YIROEVE6D5T","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"KV-Runahead: Scalable Causal LLM Inference by Parallel Key-Value Cache Generation","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.DC","authors_text":"Devang Naik, Minsik Cho, Mohammad Rastegari","submitted_at":"2024-05-08T18:03:22Z","abstract_excerpt":"Large Language Model or LLM inference has two phases, the prompt (or prefill) phase to output the first token and the extension (or decoding) phase to the generate subsequent tokens. In this work, we propose an efficient parallelization scheme, KV-Runahead to accelerate the prompt phase. The key observation is that the extension phase generates tokens faster than the prompt phase because of key-value cache (KV-cache). Hence, KV-Runahead parallelizes the prompt phase by orchestrating multiple processes to populate the KV-cache and minimizes the time-to-first-token (TTFT). Dual-purposing the KV-"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.05329","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/2405.05329/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":"2026-07-05T08:18:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"aF15Q7DxxwXUkBf3J/rnwJ18fCTbZCzNLafqI8Epcjbff+wza9091agxsvyEpYOh2CR4UFMvmceao1tRSek0Dw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T14:49:46.623467Z"},"content_sha256":"1708ddbc80b8bd5df72bda887b62fde0900f5be13197f4e70859e61708354ba6","schema_version":"1.0","event_id":"sha256:1708ddbc80b8bd5df72bda887b62fde0900f5be13197f4e70859e61708354ba6"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/IDJAH5RTFLKMAA4YIROEVE6D5T/bundle.json","state_url":"https://pith.science/pith/IDJAH5RTFLKMAA4YIROEVE6D5T/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/IDJAH5RTFLKMAA4YIROEVE6D5T/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":"2026-08-18T14:49:46Z","links":{"resolver":"https://pith.science/pith/IDJAH5RTFLKMAA4YIROEVE6D5T","bundle":"https://pith.science/pith/IDJAH5RTFLKMAA4YIROEVE6D5T/bundle.json","state":"https://pith.science/pith/IDJAH5RTFLKMAA4YIROEVE6D5T/state.json","well_known_bundle":"https://pith.science/.well-known/pith/IDJAH5RTFLKMAA4YIROEVE6D5T/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:IDJAH5RTFLKMAA4YIROEVE6D5T","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":"b0ba9ea19707a44dbc6651c023e3a7d0d584e66deb33d77c8e110c5a61336bdd","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.DC","submitted_at":"2024-05-08T18:03:22Z","title_canon_sha256":"8d5ac19dc47aed5ea37529ce7eba8339073cf99f68a891cf2c451eb388d8ed42"},"schema_version":"1.0","source":{"id":"2405.05329","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.05329","created_at":"2026-07-05T08:18:52Z"},{"alias_kind":"arxiv_version","alias_value":"2405.05329v2","created_at":"2026-07-05T08:18:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.05329","created_at":"2026-07-05T08:18:52Z"},{"alias_kind":"pith_short_12","alias_value":"IDJAH5RTFLKM","created_at":"2026-07-05T08:18:52Z"},{"alias_kind":"pith_short_16","alias_value":"IDJAH5RTFLKMAA4Y","created_at":"2026-07-05T08:18:52Z"},{"alias_kind":"pith_short_8","alias_value":"IDJAH5RT","created_at":"2026-07-05T08:18:52Z"}],"graph_snapshots":[{"event_id":"sha256:1708ddbc80b8bd5df72bda887b62fde0900f5be13197f4e70859e61708354ba6","target":"graph","created_at":"2026-07-05T08:18:52Z","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/2405.05329/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Model or LLM inference has two phases, the prompt (or prefill) phase to output the first token and the extension (or decoding) phase to the generate subsequent tokens. In this work, we propose an efficient parallelization scheme, KV-Runahead to accelerate the prompt phase. The key observation is that the extension phase generates tokens faster than the prompt phase because of key-value cache (KV-cache). Hence, KV-Runahead parallelizes the prompt phase by orchestrating multiple processes to populate the KV-cache and minimizes the time-to-first-token (TTFT). Dual-purposing the KV-","authors_text":"Devang Naik, Minsik Cho, Mohammad Rastegari","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.DC","submitted_at":"2024-05-08T18:03:22Z","title":"KV-Runahead: Scalable Causal LLM Inference by Parallel Key-Value Cache Generation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.05329","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":"sha256:7d17bf4afba3f4d14ff2dd3c68c20b8c3c3878ee8fdc97d7a59329a8dbfaa4a3","target":"record","created_at":"2026-07-05T08:18:52Z","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":"b0ba9ea19707a44dbc6651c023e3a7d0d584e66deb33d77c8e110c5a61336bdd","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.DC","submitted_at":"2024-05-08T18:03:22Z","title_canon_sha256":"8d5ac19dc47aed5ea37529ce7eba8339073cf99f68a891cf2c451eb388d8ed42"},"schema_version":"1.0","source":{"id":"2405.05329","kind":"arxiv","version":2}},"canonical_sha256":"40d203f6332ad4c00398445c4a93c3ecd75df50a3ae8e242a096e038ede6d596","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"40d203f6332ad4c00398445c4a93c3ecd75df50a3ae8e242a096e038ede6d596","first_computed_at":"2026-07-05T08:18:52.754188Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:18:52.754188Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"SizCOM9dO+1F3xh4dhu2jYnIAXswMjQ5V04ZhPFTH1bVptt9t4Nq1NGcXqfmx6yFP22tECBYVKdAlm0RvAaaDw==","signature_status":"signed_v1","signed_at":"2026-07-05T08:18:52.754625Z","signed_message":"canonical_sha256_bytes"},"source_id":"2405.05329","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7d17bf4afba3f4d14ff2dd3c68c20b8c3c3878ee8fdc97d7a59329a8dbfaa4a3","sha256:1708ddbc80b8bd5df72bda887b62fde0900f5be13197f4e70859e61708354ba6"],"state_sha256":"8aa43756deb48e5a4cf86bfdfba530e03b617b9bb5c822c859e199dd819604c1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Fur+jZYk3PVkXD0vG+8y3FYXa5gusTGk3zJ5n6y3HrwHwT8cD8LYkWHBifxXz9Jz4NlIakEfwO+zfLRGBuU7BQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-18T14:49:46.626941Z","bundle_sha256":"6f76967812d00d016c0b64b8a90aebbeb73ec88ae0ab60fe92f02cf2f2964210"}}