{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:3OEDGC2JZPPPPYWMMNJYNIHUQ4","short_pith_number":"pith:3OEDGC2J","canonical_record":{"source":{"id":"2401.09135","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-01-17T11:17:04Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"9b8250bbc58470474cd0b5405fa715a7830474d31afa09cf2b810efe350cbb22","abstract_canon_sha256":"1e9ba45fe3f6da4e828a06d51154cbe6c91617da31d25c2833e567f1546bf691"},"schema_version":"1.0"},"canonical_sha256":"db88330b49cbdef7e2cc635386a0f48733a2db353efa3c860306f6e49771f563","source":{"kind":"arxiv","id":"2401.09135","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2401.09135","created_at":"2026-07-05T09:10:30Z"},{"alias_kind":"arxiv_version","alias_value":"2401.09135v2","created_at":"2026-07-05T09:10:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.09135","created_at":"2026-07-05T09:10:30Z"},{"alias_kind":"pith_short_12","alias_value":"3OEDGC2JZPPP","created_at":"2026-07-05T09:10:30Z"},{"alias_kind":"pith_short_16","alias_value":"3OEDGC2JZPPPPYWM","created_at":"2026-07-05T09:10:30Z"},{"alias_kind":"pith_short_8","alias_value":"3OEDGC2J","created_at":"2026-07-05T09:10:30Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:3OEDGC2JZPPPPYWMMNJYNIHUQ4","target":"record","payload":{"canonical_record":{"source":{"id":"2401.09135","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-01-17T11:17:04Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"9b8250bbc58470474cd0b5405fa715a7830474d31afa09cf2b810efe350cbb22","abstract_canon_sha256":"1e9ba45fe3f6da4e828a06d51154cbe6c91617da31d25c2833e567f1546bf691"},"schema_version":"1.0"},"canonical_sha256":"db88330b49cbdef7e2cc635386a0f48733a2db353efa3c860306f6e49771f563","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:10:30.024887Z","signature_b64":"XSGHBgNSNz7vy4CmzfbhMVsOWPGHiuPZX2hKl5HgVop54aKH91RnQlFReYRT0qcb7FDSRm383tty6HaWyoC+Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"db88330b49cbdef7e2cc635386a0f48733a2db353efa3c860306f6e49771f563","last_reissued_at":"2026-07-05T09:10:30.024408Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:10:30.024408Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2401.09135","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":"2026-07-05T09:10:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+vLL7iDDnHh5Mx1cjRJNWKr0IzZY64/B4Y5HH7NO07rbpOdCMl6wU3IyAgtHuSZyBUoFXcd507x2pZFBOb62Ag==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T19:34:40.271366Z"},"content_sha256":"3120c255c46e67ee77bf1c9177a762c007dbc2d0418ffb0a5b4d573f205761ca","schema_version":"1.0","event_id":"sha256:3120c255c46e67ee77bf1c9177a762c007dbc2d0418ffb0a5b4d573f205761ca"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:3OEDGC2JZPPPPYWMMNJYNIHUQ4","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Asynchronous Local-SGD Training for Language Modeling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Andrei A. Rusu, Arthur Douillard, Arthur Szlam, Bo Liu, Jiajun Shen, Marc'Aurelio Ranzato, Rachita Chhaparia, Satyen Kale","submitted_at":"2024-01-17T11:17:04Z","abstract_excerpt":"Local stochastic gradient descent (Local-SGD), also referred to as federated averaging, is an approach to distributed optimization where each device performs more than one SGD update per communication. This work presents an empirical study of {\\it asynchronous} Local-SGD for training language models; that is, each worker updates the global parameters as soon as it has finished its SGD steps. We conduct a comprehensive investigation by examining how worker hardware heterogeneity, model size, number of workers, and optimizer could impact the learning performance. We find that with naive implemen"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.09135","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/2401.09135/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":"2026-07-05T09:10:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qYTy7ZHm8FAi/inaaLIkvV/pBmhUldD1Kihu47MrE0Za9niDKfeM8JJNNqjZtpMgnI2qndKugkMfe3BT1C2MDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T19:34:40.272355Z"},"content_sha256":"46ead5dca7e16e1329d206f6b7df33041f9d71f06f59ec208877c9a64a241ec3","schema_version":"1.0","event_id":"sha256:46ead5dca7e16e1329d206f6b7df33041f9d71f06f59ec208877c9a64a241ec3"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/3OEDGC2JZPPPPYWMMNJYNIHUQ4/bundle.json","state_url":"https://pith.science/pith/3OEDGC2JZPPPPYWMMNJYNIHUQ4/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/3OEDGC2JZPPPPYWMMNJYNIHUQ4/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":"2026-08-09T19:34:40Z","links":{"resolver":"https://pith.science/pith/3OEDGC2JZPPPPYWMMNJYNIHUQ4","bundle":"https://pith.science/pith/3OEDGC2JZPPPPYWMMNJYNIHUQ4/bundle.json","state":"https://pith.science/pith/3OEDGC2JZPPPPYWMMNJYNIHUQ4/state.json","well_known_bundle":"https://pith.science/.well-known/pith/3OEDGC2JZPPPPYWMMNJYNIHUQ4/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:3OEDGC2JZPPPPYWMMNJYNIHUQ4","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":"1e9ba45fe3f6da4e828a06d51154cbe6c91617da31d25c2833e567f1546bf691","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-01-17T11:17:04Z","title_canon_sha256":"9b8250bbc58470474cd0b5405fa715a7830474d31afa09cf2b810efe350cbb22"},"schema_version":"1.0","source":{"id":"2401.09135","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2401.09135","created_at":"2026-07-05T09:10:30Z"},{"alias_kind":"arxiv_version","alias_value":"2401.09135v2","created_at":"2026-07-05T09:10:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.09135","created_at":"2026-07-05T09:10:30Z"},{"alias_kind":"pith_short_12","alias_value":"3OEDGC2JZPPP","created_at":"2026-07-05T09:10:30Z"},{"alias_kind":"pith_short_16","alias_value":"3OEDGC2JZPPPPYWM","created_at":"2026-07-05T09:10:30Z"},{"alias_kind":"pith_short_8","alias_value":"3OEDGC2J","created_at":"2026-07-05T09:10:30Z"}],"graph_snapshots":[{"event_id":"sha256:46ead5dca7e16e1329d206f6b7df33041f9d71f06f59ec208877c9a64a241ec3","target":"graph","created_at":"2026-07-05T09:10:30Z","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/2401.09135/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Local stochastic gradient descent (Local-SGD), also referred to as federated averaging, is an approach to distributed optimization where each device performs more than one SGD update per communication. This work presents an empirical study of {\\it asynchronous} Local-SGD for training language models; that is, each worker updates the global parameters as soon as it has finished its SGD steps. We conduct a comprehensive investigation by examining how worker hardware heterogeneity, model size, number of workers, and optimizer could impact the learning performance. We find that with naive implemen","authors_text":"Andrei A. Rusu, Arthur Douillard, Arthur Szlam, Bo Liu, Jiajun Shen, Marc'Aurelio Ranzato, Rachita Chhaparia, Satyen Kale","cross_cats":["cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-01-17T11:17:04Z","title":"Asynchronous Local-SGD Training for Language Modeling"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.09135","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":"sha256:3120c255c46e67ee77bf1c9177a762c007dbc2d0418ffb0a5b4d573f205761ca","target":"record","created_at":"2026-07-05T09:10:30Z","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":"1e9ba45fe3f6da4e828a06d51154cbe6c91617da31d25c2833e567f1546bf691","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-01-17T11:17:04Z","title_canon_sha256":"9b8250bbc58470474cd0b5405fa715a7830474d31afa09cf2b810efe350cbb22"},"schema_version":"1.0","source":{"id":"2401.09135","kind":"arxiv","version":2}},"canonical_sha256":"db88330b49cbdef7e2cc635386a0f48733a2db353efa3c860306f6e49771f563","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"db88330b49cbdef7e2cc635386a0f48733a2db353efa3c860306f6e49771f563","first_computed_at":"2026-07-05T09:10:30.024408Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:10:30.024408Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"XSGHBgNSNz7vy4CmzfbhMVsOWPGHiuPZX2hKl5HgVop54aKH91RnQlFReYRT0qcb7FDSRm383tty6HaWyoC+Dg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:10:30.024887Z","signed_message":"canonical_sha256_bytes"},"source_id":"2401.09135","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3120c255c46e67ee77bf1c9177a762c007dbc2d0418ffb0a5b4d573f205761ca","sha256:46ead5dca7e16e1329d206f6b7df33041f9d71f06f59ec208877c9a64a241ec3"],"state_sha256":"a89dc6925b4c3fb2bcff300018b241040f77478c894c9c872209816af4698515"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FVLIoGkk5MfuVRlRmjbbaDfW7lNgR8uUHtkZRtQu/KbtT1roqgGQaDf6nVs99BQBb4p/pv3YFWkXe8qfnc/aAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T19:34:40.278458Z","bundle_sha256":"5ce6ae1646c51b6262a2f483968830ab0aa37c71034c1bd592bf83f58f7c4526"}}