{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:VPJ4AAWWTBCVH7Q2F3ETHTFPYI","short_pith_number":"pith:VPJ4AAWW","canonical_record":{"source":{"id":"2211.05953","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.DC","submitted_at":"2022-11-11T02:00:32Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"142a546f88497f752e5a7a9fc9cae09f8b39e2899dc10967eae120f42b8bd242","abstract_canon_sha256":"261dbe7d7774077dfa91e57ee3246b589d57ec400c46187c62133269381bbdda"},"schema_version":"1.0"},"canonical_sha256":"abd3c002d6984553fe1a2ec933ccafc21f53d930c65f4f65d4f8cf97c884555f","source":{"kind":"arxiv","id":"2211.05953","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2211.05953","created_at":"2026-07-05T06:28:32Z"},{"alias_kind":"arxiv_version","alias_value":"2211.05953v2","created_at":"2026-07-05T06:28:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2211.05953","created_at":"2026-07-05T06:28:32Z"},{"alias_kind":"pith_short_12","alias_value":"VPJ4AAWWTBCV","created_at":"2026-07-05T06:28:32Z"},{"alias_kind":"pith_short_16","alias_value":"VPJ4AAWWTBCVH7Q2","created_at":"2026-07-05T06:28:32Z"},{"alias_kind":"pith_short_8","alias_value":"VPJ4AAWW","created_at":"2026-07-05T06:28:32Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:VPJ4AAWWTBCVH7Q2F3ETHTFPYI","target":"record","payload":{"canonical_record":{"source":{"id":"2211.05953","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.DC","submitted_at":"2022-11-11T02:00:32Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"142a546f88497f752e5a7a9fc9cae09f8b39e2899dc10967eae120f42b8bd242","abstract_canon_sha256":"261dbe7d7774077dfa91e57ee3246b589d57ec400c46187c62133269381bbdda"},"schema_version":"1.0"},"canonical_sha256":"abd3c002d6984553fe1a2ec933ccafc21f53d930c65f4f65d4f8cf97c884555f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:28:32.630382Z","signature_b64":"3gdb9u59HkQpjbN2hGuZNR0RkMIG2BQ5tsSonpNmCNohe0FvK9EfMTVjIneBhTvWhnXzXzOlF6Fmg5nxVqW6Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"abd3c002d6984553fe1a2ec933ccafc21f53d930c65f4f65d4f8cf97c884555f","last_reissued_at":"2026-07-05T06:28:32.629949Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:28:32.629949Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2211.05953","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":"2026-07-05T06:28:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wguHFATOWr8ZE3eUN0xTUaNHyETsj+1hLqiAo62BrB+mwxCLdCeTW0UfXLST/Cu2stYcjjclUrGO6TZzcrYcDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-20T09:45:43.855175Z"},"content_sha256":"de928df136379249841b459535d710b55ee9fd0dbe2088ff6c8a3933d49e5667","schema_version":"1.0","event_id":"sha256:de928df136379249841b459535d710b55ee9fd0dbe2088ff6c8a3933d49e5667"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:VPJ4AAWWTBCVH7Q2F3ETHTFPYI","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Breadth-First Pipeline Parallelism","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.DC","authors_text":"Joel Lamy-Poirier","submitted_at":"2022-11-11T02:00:32Z","abstract_excerpt":"We introduce Breadth-First Pipeline Parallelism, a novel training schedule which optimizes the combination of pipeline and data parallelism. Breadth-First Pipeline Parallelism lowers training time, cost and memory usage by combining a high GPU utilization with a small batch size per GPU, and by making use of fully sharded data parallelism. Experimentally, we observed an increase of up to 43% in training throughput for a 52 billion-parameter model using a small batch size per GPU compared to Megatron-LM, which would reduce the training time and cost by the same amount on a large GPU cluster."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2211.05953","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/2211.05953/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":"2026-07-05T06:28:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7DIZ1M46eP1vvZCZF5pVCCqhU8NBFAOT88Zn89pKQFjpLTeNKrHgGyF1vE0h//4Xg/7KMJrzgm9JnxHL+1s3Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-20T09:45:43.855826Z"},"content_sha256":"90fd1ca22d2726914ca0ac4f6b4bd53423a70651e24864f80dab62f2a0823278","schema_version":"1.0","event_id":"sha256:90fd1ca22d2726914ca0ac4f6b4bd53423a70651e24864f80dab62f2a0823278"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/VPJ4AAWWTBCVH7Q2F3ETHTFPYI/bundle.json","state_url":"https://pith.science/pith/VPJ4AAWWTBCVH7Q2F3ETHTFPYI/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/VPJ4AAWWTBCVH7Q2F3ETHTFPYI/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":"2026-08-20T09:45:43Z","links":{"resolver":"https://pith.science/pith/VPJ4AAWWTBCVH7Q2F3ETHTFPYI","bundle":"https://pith.science/pith/VPJ4AAWWTBCVH7Q2F3ETHTFPYI/bundle.json","state":"https://pith.science/pith/VPJ4AAWWTBCVH7Q2F3ETHTFPYI/state.json","well_known_bundle":"https://pith.science/.well-known/pith/VPJ4AAWWTBCVH7Q2F3ETHTFPYI/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:VPJ4AAWWTBCVH7Q2F3ETHTFPYI","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":"261dbe7d7774077dfa91e57ee3246b589d57ec400c46187c62133269381bbdda","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.DC","submitted_at":"2022-11-11T02:00:32Z","title_canon_sha256":"142a546f88497f752e5a7a9fc9cae09f8b39e2899dc10967eae120f42b8bd242"},"schema_version":"1.0","source":{"id":"2211.05953","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2211.05953","created_at":"2026-07-05T06:28:32Z"},{"alias_kind":"arxiv_version","alias_value":"2211.05953v2","created_at":"2026-07-05T06:28:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2211.05953","created_at":"2026-07-05T06:28:32Z"},{"alias_kind":"pith_short_12","alias_value":"VPJ4AAWWTBCV","created_at":"2026-07-05T06:28:32Z"},{"alias_kind":"pith_short_16","alias_value":"VPJ4AAWWTBCVH7Q2","created_at":"2026-07-05T06:28:32Z"},{"alias_kind":"pith_short_8","alias_value":"VPJ4AAWW","created_at":"2026-07-05T06:28:32Z"}],"graph_snapshots":[{"event_id":"sha256:90fd1ca22d2726914ca0ac4f6b4bd53423a70651e24864f80dab62f2a0823278","target":"graph","created_at":"2026-07-05T06:28:32Z","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/2211.05953/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce Breadth-First Pipeline Parallelism, a novel training schedule which optimizes the combination of pipeline and data parallelism. Breadth-First Pipeline Parallelism lowers training time, cost and memory usage by combining a high GPU utilization with a small batch size per GPU, and by making use of fully sharded data parallelism. Experimentally, we observed an increase of up to 43% in training throughput for a 52 billion-parameter model using a small batch size per GPU compared to Megatron-LM, which would reduce the training time and cost by the same amount on a large GPU cluster.","authors_text":"Joel Lamy-Poirier","cross_cats":["cs.AI","cs.CL","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.DC","submitted_at":"2022-11-11T02:00:32Z","title":"Breadth-First Pipeline Parallelism"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2211.05953","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:de928df136379249841b459535d710b55ee9fd0dbe2088ff6c8a3933d49e5667","target":"record","created_at":"2026-07-05T06:28:32Z","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":"261dbe7d7774077dfa91e57ee3246b589d57ec400c46187c62133269381bbdda","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.DC","submitted_at":"2022-11-11T02:00:32Z","title_canon_sha256":"142a546f88497f752e5a7a9fc9cae09f8b39e2899dc10967eae120f42b8bd242"},"schema_version":"1.0","source":{"id":"2211.05953","kind":"arxiv","version":2}},"canonical_sha256":"abd3c002d6984553fe1a2ec933ccafc21f53d930c65f4f65d4f8cf97c884555f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"abd3c002d6984553fe1a2ec933ccafc21f53d930c65f4f65d4f8cf97c884555f","first_computed_at":"2026-07-05T06:28:32.629949Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:28:32.629949Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"3gdb9u59HkQpjbN2hGuZNR0RkMIG2BQ5tsSonpNmCNohe0FvK9EfMTVjIneBhTvWhnXzXzOlF6Fmg5nxVqW6Dg==","signature_status":"signed_v1","signed_at":"2026-07-05T06:28:32.630382Z","signed_message":"canonical_sha256_bytes"},"source_id":"2211.05953","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:de928df136379249841b459535d710b55ee9fd0dbe2088ff6c8a3933d49e5667","sha256:90fd1ca22d2726914ca0ac4f6b4bd53423a70651e24864f80dab62f2a0823278"],"state_sha256":"228b5eec8105ab2220d08698c22b71c31dea347baf33d111320b5e31d2a7ac98"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"I1Mo6JMI4EW6Etw9ANEY303WQXVXixRSbpiT5XrnehZLDOgMyWQV4CpAZf6LzxEin3mjDYI5jFj9xBvIH1MXDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-20T09:45:43.860016Z","bundle_sha256":"95688df27f8291ddef8332ae279ecc82f4c97ac203da6047ac71173015e2fd51"}}