{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:EJJXK7XL3N7M7GDWBDZ7MLPYDM","short_pith_number":"pith:EJJXK7XL","canonical_record":{"source":{"id":"2508.06526","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2025-08-02T03:50:14Z","cross_cats_sorted":["cs.AI","cs.AR"],"title_canon_sha256":"da9bfb659a9835ada44123f57de71515b791b47b4617e3929d60cf24afa57bd7","abstract_canon_sha256":"5ca7ca2ff300b21e13c5575d68cf85e2a674b040b513885ff8a2d3279ee9202a"},"schema_version":"1.0"},"canonical_sha256":"2253757eebdb7ecf987608f3f62df81b0a27fa4fb3f63a94c225875b5454ed31","source":{"kind":"arxiv","id":"2508.06526","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.06526","created_at":"2026-05-20T02:05:34Z"},{"alias_kind":"arxiv_version","alias_value":"2508.06526v3","created_at":"2026-05-20T02:05:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.06526","created_at":"2026-05-20T02:05:34Z"},{"alias_kind":"pith_short_12","alias_value":"EJJXK7XL3N7M","created_at":"2026-05-20T02:05:34Z"},{"alias_kind":"pith_short_16","alias_value":"EJJXK7XL3N7M7GDW","created_at":"2026-05-20T02:05:34Z"},{"alias_kind":"pith_short_8","alias_value":"EJJXK7XL","created_at":"2026-05-20T02:05:34Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:EJJXK7XL3N7M7GDWBDZ7MLPYDM","target":"record","payload":{"canonical_record":{"source":{"id":"2508.06526","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2025-08-02T03:50:14Z","cross_cats_sorted":["cs.AI","cs.AR"],"title_canon_sha256":"da9bfb659a9835ada44123f57de71515b791b47b4617e3929d60cf24afa57bd7","abstract_canon_sha256":"5ca7ca2ff300b21e13c5575d68cf85e2a674b040b513885ff8a2d3279ee9202a"},"schema_version":"1.0"},"canonical_sha256":"2253757eebdb7ecf987608f3f62df81b0a27fa4fb3f63a94c225875b5454ed31","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T02:05:34.663633Z","signature_b64":"zayKa/WxjLF2xA6N0oy+XY/byOt07EVordI03oN/oFOMIWJGhhmh3EARhHMC/EcqE08aGosvNz7BRMslZSg9AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2253757eebdb7ecf987608f3f62df81b0a27fa4fb3f63a94c225875b5454ed31","last_reissued_at":"2026-05-20T02:05:34.662750Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T02:05:34.662750Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2508.06526","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T02:05:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"p+JwPXgJpL973j2S9cIS32GmYyDKtzGG0Um8BnVvr2LvTWVuH39QqtT0Hd9h+fJJJXxrnjGAhJNwwQUnNQsoAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T02:35:21.646372Z"},"content_sha256":"d70bc06411e138321674ed4c79e19fb71cc94695d08ea1bc60026e3a76ab0f2d","schema_version":"1.0","event_id":"sha256:d70bc06411e138321674ed4c79e19fb71cc94695d08ea1bc60026e3a76ab0f2d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:EJJXK7XL3N7M7GDWBDZ7MLPYDM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"PiKV: KV Cache Management System for Mixture of Experts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.AR"],"primary_cat":"cs.DC","authors_text":"Ben Lengerich, Dong Liu, Yanxuan Yu, Ying Nian Wu","submitted_at":"2025-08-02T03:50:14Z","abstract_excerpt":"As large-scale language models continue to scale up in both size and context length, the memory and communication cost of key-value (KV) cache storage has become a major bottleneck in multi-GPU and multi-node inference. While MoE-based architectures sparsify computation across experts, the corresponding KV caches remain dense and globally synchronized, resulting in significant overhead.\n  We introduce \\textbf{PiKV}, a parallel and distributed KV cache serving framework tailored for MoE architecture. PiKV leverages \\textit{expert-sharded KV storage} to partition caches across GPUs, \\textit{PiKV"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.06526","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/2508.06526/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":"2026-05-20T02:05:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5RzDTnYyDFtnjQpb38rAiUCdM13YJOD8Otpr5WZ8bIZkxfsKZyGbEdSaNhNNoVcdYmcV9kF9xGFlIBiPzPYjCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T02:35:21.647063Z"},"content_sha256":"39203aec52fa184294adf9ab0f7523fd7d531558c00f9b7fab6985e0129087e5","schema_version":"1.0","event_id":"sha256:39203aec52fa184294adf9ab0f7523fd7d531558c00f9b7fab6985e0129087e5"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/EJJXK7XL3N7M7GDWBDZ7MLPYDM/bundle.json","state_url":"https://pith.science/pith/EJJXK7XL3N7M7GDWBDZ7MLPYDM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/EJJXK7XL3N7M7GDWBDZ7MLPYDM/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":"2026-05-27T02:35:21Z","links":{"resolver":"https://pith.science/pith/EJJXK7XL3N7M7GDWBDZ7MLPYDM","bundle":"https://pith.science/pith/EJJXK7XL3N7M7GDWBDZ7MLPYDM/bundle.json","state":"https://pith.science/pith/EJJXK7XL3N7M7GDWBDZ7MLPYDM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/EJJXK7XL3N7M7GDWBDZ7MLPYDM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:EJJXK7XL3N7M7GDWBDZ7MLPYDM","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":"5ca7ca2ff300b21e13c5575d68cf85e2a674b040b513885ff8a2d3279ee9202a","cross_cats_sorted":["cs.AI","cs.AR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2025-08-02T03:50:14Z","title_canon_sha256":"da9bfb659a9835ada44123f57de71515b791b47b4617e3929d60cf24afa57bd7"},"schema_version":"1.0","source":{"id":"2508.06526","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.06526","created_at":"2026-05-20T02:05:34Z"},{"alias_kind":"arxiv_version","alias_value":"2508.06526v3","created_at":"2026-05-20T02:05:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.06526","created_at":"2026-05-20T02:05:34Z"},{"alias_kind":"pith_short_12","alias_value":"EJJXK7XL3N7M","created_at":"2026-05-20T02:05:34Z"},{"alias_kind":"pith_short_16","alias_value":"EJJXK7XL3N7M7GDW","created_at":"2026-05-20T02:05:34Z"},{"alias_kind":"pith_short_8","alias_value":"EJJXK7XL","created_at":"2026-05-20T02:05:34Z"}],"graph_snapshots":[{"event_id":"sha256:39203aec52fa184294adf9ab0f7523fd7d531558c00f9b7fab6985e0129087e5","target":"graph","created_at":"2026-05-20T02:05:34Z","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/2508.06526/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As large-scale language models continue to scale up in both size and context length, the memory and communication cost of key-value (KV) cache storage has become a major bottleneck in multi-GPU and multi-node inference. While MoE-based architectures sparsify computation across experts, the corresponding KV caches remain dense and globally synchronized, resulting in significant overhead.\n  We introduce \\textbf{PiKV}, a parallel and distributed KV cache serving framework tailored for MoE architecture. PiKV leverages \\textit{expert-sharded KV storage} to partition caches across GPUs, \\textit{PiKV","authors_text":"Ben Lengerich, Dong Liu, Yanxuan Yu, Ying Nian Wu","cross_cats":["cs.AI","cs.AR"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2025-08-02T03:50:14Z","title":"PiKV: KV Cache Management System for Mixture of Experts"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.06526","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:d70bc06411e138321674ed4c79e19fb71cc94695d08ea1bc60026e3a76ab0f2d","target":"record","created_at":"2026-05-20T02:05:34Z","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":"5ca7ca2ff300b21e13c5575d68cf85e2a674b040b513885ff8a2d3279ee9202a","cross_cats_sorted":["cs.AI","cs.AR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2025-08-02T03:50:14Z","title_canon_sha256":"da9bfb659a9835ada44123f57de71515b791b47b4617e3929d60cf24afa57bd7"},"schema_version":"1.0","source":{"id":"2508.06526","kind":"arxiv","version":3}},"canonical_sha256":"2253757eebdb7ecf987608f3f62df81b0a27fa4fb3f63a94c225875b5454ed31","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2253757eebdb7ecf987608f3f62df81b0a27fa4fb3f63a94c225875b5454ed31","first_computed_at":"2026-05-20T02:05:34.662750Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T02:05:34.662750Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"zayKa/WxjLF2xA6N0oy+XY/byOt07EVordI03oN/oFOMIWJGhhmh3EARhHMC/EcqE08aGosvNz7BRMslZSg9AA==","signature_status":"signed_v1","signed_at":"2026-05-20T02:05:34.663633Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.06526","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d70bc06411e138321674ed4c79e19fb71cc94695d08ea1bc60026e3a76ab0f2d","sha256:39203aec52fa184294adf9ab0f7523fd7d531558c00f9b7fab6985e0129087e5"],"state_sha256":"ffd576e2766b9d37cf4f4b9a98adfdf5d2a7730391a877a0b4d2e01a814850f2"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7KRhDgJfwU46+/IUsZRMjT2P/dx2JKs8wtd161Ce6//DhplE4a45YFBJP4EoZtOv2t/bsiYk7Ecoy8XHxs9SCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-27T02:35:21.650838Z","bundle_sha256":"c6053476194b09bd994fa3f49e8ecb4f2376c7d7cba27b4a29daf4611781d57c"}}