{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:OKGUPZFP62ETLNBRA3EMFXH5PH","short_pith_number":"pith:OKGUPZFP","canonical_record":{"source":{"id":"2411.17089","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2024-11-26T04:03:14Z","cross_cats_sorted":["cs.DC","cs.PF"],"title_canon_sha256":"41f3f0255e68c3461b8e7e65f02b73487fc09411bb544725fae42bf8992bf148","abstract_canon_sha256":"15379c7bddb190a4a2cbc33195ee83c61c23c3c13e75dee4e38ca69b0f15951f"},"schema_version":"1.0"},"canonical_sha256":"728d47e4aff68935b43106c8c2dcfd79f9ed4252119bea0b9f727fe152869480","source":{"kind":"arxiv","id":"2411.17089","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.17089","created_at":"2026-07-05T11:15:40Z"},{"alias_kind":"arxiv_version","alias_value":"2411.17089v2","created_at":"2026-07-05T11:15:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.17089","created_at":"2026-07-05T11:15:40Z"},{"alias_kind":"pith_short_12","alias_value":"OKGUPZFP62ET","created_at":"2026-07-05T11:15:40Z"},{"alias_kind":"pith_short_16","alias_value":"OKGUPZFP62ETLNBR","created_at":"2026-07-05T11:15:40Z"},{"alias_kind":"pith_short_8","alias_value":"OKGUPZFP","created_at":"2026-07-05T11:15:40Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:OKGUPZFP62ETLNBRA3EMFXH5PH","target":"record","payload":{"canonical_record":{"source":{"id":"2411.17089","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2024-11-26T04:03:14Z","cross_cats_sorted":["cs.DC","cs.PF"],"title_canon_sha256":"41f3f0255e68c3461b8e7e65f02b73487fc09411bb544725fae42bf8992bf148","abstract_canon_sha256":"15379c7bddb190a4a2cbc33195ee83c61c23c3c13e75dee4e38ca69b0f15951f"},"schema_version":"1.0"},"canonical_sha256":"728d47e4aff68935b43106c8c2dcfd79f9ed4252119bea0b9f727fe152869480","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:15:40.695683Z","signature_b64":"VlODS6KewFBJjOcsHmXCoM2FNl4Ghxyvol9eBNK4AkXvwyglyC8PAoHNlZ4PGLowYwkgGRvL5qiZ08dQ3r+oCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"728d47e4aff68935b43106c8c2dcfd79f9ed4252119bea0b9f727fe152869480","last_reissued_at":"2026-07-05T11:15:40.695203Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:15:40.695203Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2411.17089","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":"2026-07-05T11:15:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JwEDvWg7WHPOd/OFB13vIiKPL2X6ZCfQez9PEtHdu2cwQzG01Nkgm0WH//N/1i2U2dv/aaBg+knFDkUi7zAYAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T11:10:29.237103Z"},"content_sha256":"ecc582514a7be2eaf1fe27484ba7efb733e2b2813e03da5cf0776e28c5c88952","schema_version":"1.0","event_id":"sha256:ecc582514a7be2eaf1fe27484ba7efb733e2b2813e03da5cf0776e28c5c88952"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:OKGUPZFP62ETLNBRA3EMFXH5PH","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"KVPR: Efficient LLM Inference with I/O-Aware KV Cache Partial Recomputation","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.DC","cs.PF"],"primary_cat":"cs.LG","authors_text":"Chaoyi Jiang, Hossein Entezari Zarch, Lei Gao, Murali Annavaram","submitted_at":"2024-11-26T04:03:14Z","abstract_excerpt":"Inference for Large Language Models (LLMs) is computationally demanding. To reduce the cost of auto-regressive decoding, Key-Value (KV) cache is used to store intermediate activations, which significantly lowers the computational overhead for token generation. However, the memory required for the KV cache grows rapidly, often exceeding the capacity of GPU memory. A cost-effective alternative is to offload KV cache to CPU memory, which alleviates GPU memory pressure, but shifts the bottleneck to the limited bandwidth of the PCIe connection between the CPU and GPU. Existing methods attempt to ad"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.17089","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/2411.17089/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":"2026-07-05T11:15:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"W5ckGWTS3Ujq2blYVoS5/zgGSrVsDYRAOCq9tv7yNJrIpcgaDct0eUyb+jhJguDkqOUWdMws8a1MBswYRHa0DA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T11:10:29.237830Z"},"content_sha256":"c62264be72127b786bc2a4af9ff046cb2f0831bb76c175313e795ae18cca46d9","schema_version":"1.0","event_id":"sha256:c62264be72127b786bc2a4af9ff046cb2f0831bb76c175313e795ae18cca46d9"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/OKGUPZFP62ETLNBRA3EMFXH5PH/bundle.json","state_url":"https://pith.science/pith/OKGUPZFP62ETLNBRA3EMFXH5PH/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/OKGUPZFP62ETLNBRA3EMFXH5PH/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":"2026-08-08T11:10:29Z","links":{"resolver":"https://pith.science/pith/OKGUPZFP62ETLNBRA3EMFXH5PH","bundle":"https://pith.science/pith/OKGUPZFP62ETLNBRA3EMFXH5PH/bundle.json","state":"https://pith.science/pith/OKGUPZFP62ETLNBRA3EMFXH5PH/state.json","well_known_bundle":"https://pith.science/.well-known/pith/OKGUPZFP62ETLNBRA3EMFXH5PH/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:OKGUPZFP62ETLNBRA3EMFXH5PH","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":"15379c7bddb190a4a2cbc33195ee83c61c23c3c13e75dee4e38ca69b0f15951f","cross_cats_sorted":["cs.DC","cs.PF"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2024-11-26T04:03:14Z","title_canon_sha256":"41f3f0255e68c3461b8e7e65f02b73487fc09411bb544725fae42bf8992bf148"},"schema_version":"1.0","source":{"id":"2411.17089","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.17089","created_at":"2026-07-05T11:15:40Z"},{"alias_kind":"arxiv_version","alias_value":"2411.17089v2","created_at":"2026-07-05T11:15:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.17089","created_at":"2026-07-05T11:15:40Z"},{"alias_kind":"pith_short_12","alias_value":"OKGUPZFP62ET","created_at":"2026-07-05T11:15:40Z"},{"alias_kind":"pith_short_16","alias_value":"OKGUPZFP62ETLNBR","created_at":"2026-07-05T11:15:40Z"},{"alias_kind":"pith_short_8","alias_value":"OKGUPZFP","created_at":"2026-07-05T11:15:40Z"}],"graph_snapshots":[{"event_id":"sha256:c62264be72127b786bc2a4af9ff046cb2f0831bb76c175313e795ae18cca46d9","target":"graph","created_at":"2026-07-05T11:15:40Z","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/2411.17089/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Inference for Large Language Models (LLMs) is computationally demanding. To reduce the cost of auto-regressive decoding, Key-Value (KV) cache is used to store intermediate activations, which significantly lowers the computational overhead for token generation. However, the memory required for the KV cache grows rapidly, often exceeding the capacity of GPU memory. A cost-effective alternative is to offload KV cache to CPU memory, which alleviates GPU memory pressure, but shifts the bottleneck to the limited bandwidth of the PCIe connection between the CPU and GPU. Existing methods attempt to ad","authors_text":"Chaoyi Jiang, Hossein Entezari Zarch, Lei Gao, Murali Annavaram","cross_cats":["cs.DC","cs.PF"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2024-11-26T04:03:14Z","title":"KVPR: Efficient LLM Inference with I/O-Aware KV Cache Partial Recomputation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.17089","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:ecc582514a7be2eaf1fe27484ba7efb733e2b2813e03da5cf0776e28c5c88952","target":"record","created_at":"2026-07-05T11:15:40Z","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":"15379c7bddb190a4a2cbc33195ee83c61c23c3c13e75dee4e38ca69b0f15951f","cross_cats_sorted":["cs.DC","cs.PF"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2024-11-26T04:03:14Z","title_canon_sha256":"41f3f0255e68c3461b8e7e65f02b73487fc09411bb544725fae42bf8992bf148"},"schema_version":"1.0","source":{"id":"2411.17089","kind":"arxiv","version":2}},"canonical_sha256":"728d47e4aff68935b43106c8c2dcfd79f9ed4252119bea0b9f727fe152869480","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"728d47e4aff68935b43106c8c2dcfd79f9ed4252119bea0b9f727fe152869480","first_computed_at":"2026-07-05T11:15:40.695203Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:15:40.695203Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"VlODS6KewFBJjOcsHmXCoM2FNl4Ghxyvol9eBNK4AkXvwyglyC8PAoHNlZ4PGLowYwkgGRvL5qiZ08dQ3r+oCA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:15:40.695683Z","signed_message":"canonical_sha256_bytes"},"source_id":"2411.17089","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ecc582514a7be2eaf1fe27484ba7efb733e2b2813e03da5cf0776e28c5c88952","sha256:c62264be72127b786bc2a4af9ff046cb2f0831bb76c175313e795ae18cca46d9"],"state_sha256":"428efb310d10e60bd2b584c706d493fc0be4ec1668fca547d376db69bff279b4"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ts6wnoRZPnbCOMebr2ErZV2SK1xZj1CPDtC9wQe1ip4T66SPcUHQVZJ0tTY7Uft3Xj8KG3OlZZuBUo8DhOTxAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T11:10:29.244546Z","bundle_sha256":"7a1b4bbe5b239f2afe50d97420181a7a11fbf74c051846061dcdef38229f8053"}}