{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:3B3D6YE7KFYMQHEBZEZFXOT72C","short_pith_number":"pith:3B3D6YE7","canonical_record":{"source":{"id":"2210.17223","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2022-10-31T11:07:23Z","cross_cats_sorted":[],"title_canon_sha256":"9a461b9f385794ef14186fcbbeca86492682073f31fd58f0a2c6f84caa27b32a","abstract_canon_sha256":"3fb952bdfa2b9eba0b8b2e3febbed1e91bc425be1da9437c85204f0eb6ab4fc9"},"schema_version":"1.0"},"canonical_sha256":"d8763f609f5170c81c81c9325bba7fd09abfcecc50bad7be8704a28e826bf72b","source":{"kind":"arxiv","id":"2210.17223","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2210.17223","created_at":"2026-07-05T08:12:41Z"},{"alias_kind":"arxiv_version","alias_value":"2210.17223v2","created_at":"2026-07-05T08:12:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.17223","created_at":"2026-07-05T08:12:41Z"},{"alias_kind":"pith_short_12","alias_value":"3B3D6YE7KFYM","created_at":"2026-07-05T08:12:41Z"},{"alias_kind":"pith_short_16","alias_value":"3B3D6YE7KFYMQHEB","created_at":"2026-07-05T08:12:41Z"},{"alias_kind":"pith_short_8","alias_value":"3B3D6YE7","created_at":"2026-07-05T08:12:41Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:3B3D6YE7KFYMQHEBZEZFXOT72C","target":"record","payload":{"canonical_record":{"source":{"id":"2210.17223","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2022-10-31T11:07:23Z","cross_cats_sorted":[],"title_canon_sha256":"9a461b9f385794ef14186fcbbeca86492682073f31fd58f0a2c6f84caa27b32a","abstract_canon_sha256":"3fb952bdfa2b9eba0b8b2e3febbed1e91bc425be1da9437c85204f0eb6ab4fc9"},"schema_version":"1.0"},"canonical_sha256":"d8763f609f5170c81c81c9325bba7fd09abfcecc50bad7be8704a28e826bf72b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:12:41.894584Z","signature_b64":"GYNA4KW+5gTDBMqK1M9g5h/yUU7Lo0nq3t+zhS7YmMAOjUmuBwxw3auF+IB4S/cm0HKWCqDGs3qzJqzgVe4zCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d8763f609f5170c81c81c9325bba7fd09abfcecc50bad7be8704a28e826bf72b","last_reissued_at":"2026-07-05T08:12:41.894108Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:12:41.894108Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2210.17223","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":"2026-07-05T08:12:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8g0ws/FsDTgmOlP9NaMDoizLoOLSXStc7T8MZqqoqnhnCbNIahWBlo7fJmZJivH3nvwdW3fdqBRqTct8PsjSDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-16T23:34:37.508511Z"},"content_sha256":"2adc1554796ba44ec71c445249defc87aef21ce5166aa762e9ecb52602c999a7","schema_version":"1.0","event_id":"sha256:2adc1554796ba44ec71c445249defc87aef21ce5166aa762e9ecb52602c999a7"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:3B3D6YE7KFYMQHEBZEZFXOT72C","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Accelerating Distributed MoE Training and Inference with Lina","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Cong Wang, Hong Xu, Jiamin Li, Yibo Zhu, Yimin Jiang","submitted_at":"2022-10-31T11:07:23Z","abstract_excerpt":"Scaling model parameters improves model quality at the price of high computation overhead. Sparsely activated models, usually in the form of Mixture of Experts (MoE) architecture, have sub-linear scaling of computation cost with model size, thus providing opportunities to train and serve a larger model at lower cost than their dense counterparts. However, distributed MoE training and inference is inefficient, mainly due to the interleaved all-to-all communication during model computation. This paper makes two main contributions. First, we systematically analyze all-to-all overhead in distribut"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.17223","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/2210.17223/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":"2026-07-05T08:12:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Kfo0QtXaO18f2GyGQctUw6VrafKEr7d/hre0X1jH8U4QIWRNa3THmxXW/3ZQLWnuoDUYFr6CBTzsinzfVGfCDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-16T23:34:37.509432Z"},"content_sha256":"4df527b00e2b8859def67247f20f6767c02ce38cf2566ec83c32cb6a7ffd264f","schema_version":"1.0","event_id":"sha256:4df527b00e2b8859def67247f20f6767c02ce38cf2566ec83c32cb6a7ffd264f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/3B3D6YE7KFYMQHEBZEZFXOT72C/bundle.json","state_url":"https://pith.science/pith/3B3D6YE7KFYMQHEBZEZFXOT72C/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/3B3D6YE7KFYMQHEBZEZFXOT72C/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":"2026-08-16T23:34:37Z","links":{"resolver":"https://pith.science/pith/3B3D6YE7KFYMQHEBZEZFXOT72C","bundle":"https://pith.science/pith/3B3D6YE7KFYMQHEBZEZFXOT72C/bundle.json","state":"https://pith.science/pith/3B3D6YE7KFYMQHEBZEZFXOT72C/state.json","well_known_bundle":"https://pith.science/.well-known/pith/3B3D6YE7KFYMQHEBZEZFXOT72C/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:3B3D6YE7KFYMQHEBZEZFXOT72C","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":"3fb952bdfa2b9eba0b8b2e3febbed1e91bc425be1da9437c85204f0eb6ab4fc9","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2022-10-31T11:07:23Z","title_canon_sha256":"9a461b9f385794ef14186fcbbeca86492682073f31fd58f0a2c6f84caa27b32a"},"schema_version":"1.0","source":{"id":"2210.17223","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2210.17223","created_at":"2026-07-05T08:12:41Z"},{"alias_kind":"arxiv_version","alias_value":"2210.17223v2","created_at":"2026-07-05T08:12:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.17223","created_at":"2026-07-05T08:12:41Z"},{"alias_kind":"pith_short_12","alias_value":"3B3D6YE7KFYM","created_at":"2026-07-05T08:12:41Z"},{"alias_kind":"pith_short_16","alias_value":"3B3D6YE7KFYMQHEB","created_at":"2026-07-05T08:12:41Z"},{"alias_kind":"pith_short_8","alias_value":"3B3D6YE7","created_at":"2026-07-05T08:12:41Z"}],"graph_snapshots":[{"event_id":"sha256:4df527b00e2b8859def67247f20f6767c02ce38cf2566ec83c32cb6a7ffd264f","target":"graph","created_at":"2026-07-05T08:12:41Z","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/2210.17223/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Scaling model parameters improves model quality at the price of high computation overhead. Sparsely activated models, usually in the form of Mixture of Experts (MoE) architecture, have sub-linear scaling of computation cost with model size, thus providing opportunities to train and serve a larger model at lower cost than their dense counterparts. However, distributed MoE training and inference is inefficient, mainly due to the interleaved all-to-all communication during model computation. This paper makes two main contributions. First, we systematically analyze all-to-all overhead in distribut","authors_text":"Cong Wang, Hong Xu, Jiamin Li, Yibo Zhu, Yimin Jiang","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2022-10-31T11:07:23Z","title":"Accelerating Distributed MoE Training and Inference with Lina"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.17223","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":"sha256:2adc1554796ba44ec71c445249defc87aef21ce5166aa762e9ecb52602c999a7","target":"record","created_at":"2026-07-05T08:12:41Z","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":"3fb952bdfa2b9eba0b8b2e3febbed1e91bc425be1da9437c85204f0eb6ab4fc9","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2022-10-31T11:07:23Z","title_canon_sha256":"9a461b9f385794ef14186fcbbeca86492682073f31fd58f0a2c6f84caa27b32a"},"schema_version":"1.0","source":{"id":"2210.17223","kind":"arxiv","version":2}},"canonical_sha256":"d8763f609f5170c81c81c9325bba7fd09abfcecc50bad7be8704a28e826bf72b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d8763f609f5170c81c81c9325bba7fd09abfcecc50bad7be8704a28e826bf72b","first_computed_at":"2026-07-05T08:12:41.894108Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:12:41.894108Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"GYNA4KW+5gTDBMqK1M9g5h/yUU7Lo0nq3t+zhS7YmMAOjUmuBwxw3auF+IB4S/cm0HKWCqDGs3qzJqzgVe4zCg==","signature_status":"signed_v1","signed_at":"2026-07-05T08:12:41.894584Z","signed_message":"canonical_sha256_bytes"},"source_id":"2210.17223","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2adc1554796ba44ec71c445249defc87aef21ce5166aa762e9ecb52602c999a7","sha256:4df527b00e2b8859def67247f20f6767c02ce38cf2566ec83c32cb6a7ffd264f"],"state_sha256":"8cb92502f1e13a0278744e4f41dd248c810d4ab0f978c38f7eaf89b16f083d0a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"QO4Wb17xokyadMjArrSMUYiwamPfWD7Yd8SA07Zi8ByJDvEz4yCT8Yg1O2LlBO6RLH3XL6SwOkYTmJ76PswhBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-16T23:34:37.515752Z","bundle_sha256":"684bb26a76e51a0c104b9c5ab319b435eaf11f7fc6092327bac804915a1f6da2"}}