{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2020:SD2XBP5OYHAVICP7Q7TXTQ72FJ","short_pith_number":"pith:SD2XBP5O","canonical_record":{"source":{"id":"2006.13484","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-06-24T05:00:41Z","cross_cats_sorted":["cs.CL","cs.DC","stat.ML"],"title_canon_sha256":"8aedb2c8d338d5e6770b332b071474b80f61ce6a5d6e23cca6b7806797ad8dae","abstract_canon_sha256":"7ddc53e4de92523392449bf4a8595ad73b8e99e072ee19a9501461f037e3e298"},"schema_version":"1.0"},"canonical_sha256":"90f570bfaec1c15409ff87e779c3fa2a4ca0f596d6a12cbd073f83a5ffef08c4","source":{"kind":"arxiv","id":"2006.13484","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2006.13484","created_at":"2026-07-05T01:36:18Z"},{"alias_kind":"arxiv_version","alias_value":"2006.13484v2","created_at":"2026-07-05T01:36:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2006.13484","created_at":"2026-07-05T01:36:18Z"},{"alias_kind":"pith_short_12","alias_value":"SD2XBP5OYHAV","created_at":"2026-07-05T01:36:18Z"},{"alias_kind":"pith_short_16","alias_value":"SD2XBP5OYHAVICP7","created_at":"2026-07-05T01:36:18Z"},{"alias_kind":"pith_short_8","alias_value":"SD2XBP5O","created_at":"2026-07-05T01:36:18Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2020:SD2XBP5OYHAVICP7Q7TXTQ72FJ","target":"record","payload":{"canonical_record":{"source":{"id":"2006.13484","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-06-24T05:00:41Z","cross_cats_sorted":["cs.CL","cs.DC","stat.ML"],"title_canon_sha256":"8aedb2c8d338d5e6770b332b071474b80f61ce6a5d6e23cca6b7806797ad8dae","abstract_canon_sha256":"7ddc53e4de92523392449bf4a8595ad73b8e99e072ee19a9501461f037e3e298"},"schema_version":"1.0"},"canonical_sha256":"90f570bfaec1c15409ff87e779c3fa2a4ca0f596d6a12cbd073f83a5ffef08c4","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:36:18.080220Z","signature_b64":"IeArbQ5KWxakUJihTU4k+rV6c2XIyv80kjuTMp5T7Mx5OygBYqJT1h8YSpukwjzJ9b1Nbp1DiUm87tH/AvMkCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"90f570bfaec1c15409ff87e779c3fa2a4ca0f596d6a12cbd073f83a5ffef08c4","last_reissued_at":"2026-07-05T01:36:18.079836Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:36:18.079836Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2006.13484","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":"2026-07-05T01:36:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"4P4QRrfLpSxYKriylF/PzhwOsolgMK7D7UB6H/tXQWsPzAVdlqWHqRegqrobR82f7dAZ/fmn5csFKS6qGILlAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T16:49:24.052279Z"},"content_sha256":"d17f6e588603343b88978984d8c156610d4b22aef3ca7a04ce659acd3592d53b","schema_version":"1.0","event_id":"sha256:d17f6e588603343b88978984d8c156610d4b22aef3ca7a04ce659acd3592d53b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2020:SD2XBP5OYHAVICP7Q7TXTQ72FJ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Accelerated Large Batch Optimization of BERT Pretraining in 54 minutes","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.DC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Haibin Lin, Mu Li, Sheng Zha, Shuai Zheng","submitted_at":"2020-06-24T05:00:41Z","abstract_excerpt":"BERT has recently attracted a lot of attention in natural language understanding (NLU) and achieved state-of-the-art results in various NLU tasks. However, its success requires large deep neural networks and huge amount of data, which result in long training time and impede development progress. Using stochastic gradient methods with large mini-batch has been advocated as an efficient tool to reduce the training time. Along this line of research, LAMB is a prominent example that reduces the training time of BERT from 3 days to 76 minutes on a TPUv3 Pod. In this paper, we propose an accelerated"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2006.13484","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/2006.13484/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":"2026-07-05T01:36:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IuzNou1qDrWbgj7Zh68aOHS3i5h8NtRui5VXiLTgo19YzfUT1aX8ZEKhw1uO0zRUifAFJPadi9oOSYMkzp9LAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T16:49:24.062072Z"},"content_sha256":"3dcf7650c2c64498e6fc3c6d1db9bbd6fc11967ff9ca22cf89b04bd3c0e2fd36","schema_version":"1.0","event_id":"sha256:3dcf7650c2c64498e6fc3c6d1db9bbd6fc11967ff9ca22cf89b04bd3c0e2fd36"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/SD2XBP5OYHAVICP7Q7TXTQ72FJ/bundle.json","state_url":"https://pith.science/pith/SD2XBP5OYHAVICP7Q7TXTQ72FJ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/SD2XBP5OYHAVICP7Q7TXTQ72FJ/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":"2026-08-07T16:49:24Z","links":{"resolver":"https://pith.science/pith/SD2XBP5OYHAVICP7Q7TXTQ72FJ","bundle":"https://pith.science/pith/SD2XBP5OYHAVICP7Q7TXTQ72FJ/bundle.json","state":"https://pith.science/pith/SD2XBP5OYHAVICP7Q7TXTQ72FJ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/SD2XBP5OYHAVICP7Q7TXTQ72FJ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2020:SD2XBP5OYHAVICP7Q7TXTQ72FJ","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":"7ddc53e4de92523392449bf4a8595ad73b8e99e072ee19a9501461f037e3e298","cross_cats_sorted":["cs.CL","cs.DC","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-06-24T05:00:41Z","title_canon_sha256":"8aedb2c8d338d5e6770b332b071474b80f61ce6a5d6e23cca6b7806797ad8dae"},"schema_version":"1.0","source":{"id":"2006.13484","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2006.13484","created_at":"2026-07-05T01:36:18Z"},{"alias_kind":"arxiv_version","alias_value":"2006.13484v2","created_at":"2026-07-05T01:36:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2006.13484","created_at":"2026-07-05T01:36:18Z"},{"alias_kind":"pith_short_12","alias_value":"SD2XBP5OYHAV","created_at":"2026-07-05T01:36:18Z"},{"alias_kind":"pith_short_16","alias_value":"SD2XBP5OYHAVICP7","created_at":"2026-07-05T01:36:18Z"},{"alias_kind":"pith_short_8","alias_value":"SD2XBP5O","created_at":"2026-07-05T01:36:18Z"}],"graph_snapshots":[{"event_id":"sha256:3dcf7650c2c64498e6fc3c6d1db9bbd6fc11967ff9ca22cf89b04bd3c0e2fd36","target":"graph","created_at":"2026-07-05T01:36:18Z","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/2006.13484/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"BERT has recently attracted a lot of attention in natural language understanding (NLU) and achieved state-of-the-art results in various NLU tasks. However, its success requires large deep neural networks and huge amount of data, which result in long training time and impede development progress. Using stochastic gradient methods with large mini-batch has been advocated as an efficient tool to reduce the training time. Along this line of research, LAMB is a prominent example that reduces the training time of BERT from 3 days to 76 minutes on a TPUv3 Pod. In this paper, we propose an accelerated","authors_text":"Haibin Lin, Mu Li, Sheng Zha, Shuai Zheng","cross_cats":["cs.CL","cs.DC","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-06-24T05:00:41Z","title":"Accelerated Large Batch Optimization of BERT Pretraining in 54 minutes"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2006.13484","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:d17f6e588603343b88978984d8c156610d4b22aef3ca7a04ce659acd3592d53b","target":"record","created_at":"2026-07-05T01:36:18Z","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":"7ddc53e4de92523392449bf4a8595ad73b8e99e072ee19a9501461f037e3e298","cross_cats_sorted":["cs.CL","cs.DC","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-06-24T05:00:41Z","title_canon_sha256":"8aedb2c8d338d5e6770b332b071474b80f61ce6a5d6e23cca6b7806797ad8dae"},"schema_version":"1.0","source":{"id":"2006.13484","kind":"arxiv","version":2}},"canonical_sha256":"90f570bfaec1c15409ff87e779c3fa2a4ca0f596d6a12cbd073f83a5ffef08c4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"90f570bfaec1c15409ff87e779c3fa2a4ca0f596d6a12cbd073f83a5ffef08c4","first_computed_at":"2026-07-05T01:36:18.079836Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T01:36:18.079836Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"IeArbQ5KWxakUJihTU4k+rV6c2XIyv80kjuTMp5T7Mx5OygBYqJT1h8YSpukwjzJ9b1Nbp1DiUm87tH/AvMkCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T01:36:18.080220Z","signed_message":"canonical_sha256_bytes"},"source_id":"2006.13484","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d17f6e588603343b88978984d8c156610d4b22aef3ca7a04ce659acd3592d53b","sha256:3dcf7650c2c64498e6fc3c6d1db9bbd6fc11967ff9ca22cf89b04bd3c0e2fd36"],"state_sha256":"7e40cc7b8242b3f1ca96a6473bc2921aad9059c4b23272202efcdb40a919a98b"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"F9xWSyZO4ZgLvSmjYN4Od9AZDLmdmWO7dYHmq9k1jFjoyKAqJFMjwmvtX8gYG63i4KshxD9tSTiDiwwB2QnUAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T16:49:24.073356Z","bundle_sha256":"a5452ef035c35b0f5a1d2b6e70ee6fc0fa2c3d9c279d9412925754a7f5afb4a1"}}