{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:6YMEZVXP3VFJYLMUUGCAWHUCQB","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":"80487bc19d6c54c99ffbee20395b4d3358f8448ecb1eff6cdf297ca8f0fa1515","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-04-15T18:17:12Z","title_canon_sha256":"919f1d0ed3526fda30801fa55e08d0efcfc82b0a5248fda1fd476929e2f5bab6"},"schema_version":"1.0","source":{"id":"2104.07705","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2104.07705","created_at":"2026-07-05T03:12:48Z"},{"alias_kind":"arxiv_version","alias_value":"2104.07705v2","created_at":"2026-07-05T03:12:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2104.07705","created_at":"2026-07-05T03:12:48Z"},{"alias_kind":"pith_short_12","alias_value":"6YMEZVXP3VFJ","created_at":"2026-07-05T03:12:48Z"},{"alias_kind":"pith_short_16","alias_value":"6YMEZVXP3VFJYLMU","created_at":"2026-07-05T03:12:48Z"},{"alias_kind":"pith_short_8","alias_value":"6YMEZVXP","created_at":"2026-07-05T03:12:48Z"}],"graph_snapshots":[{"event_id":"sha256:756f7e3b3e11270283deb7f5e3315f49d8f9fa8b7fda80fbeda41810fd0d655a","target":"graph","created_at":"2026-07-05T03:12:48Z","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/2104.07705/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"While large language models a la BERT are used ubiquitously in NLP, pretraining them is considered a luxury that only a few well-funded industry labs can afford. How can one train such models with a more modest budget? We present a recipe for pretraining a masked language model in 24 hours using a single low-end deep learning server. We demonstrate that through a combination of software optimizations, design choices, and hyperparameter tuning, it is possible to produce models that are competitive with BERT-base on GLUE tasks at a fraction of the original pretraining cost.","authors_text":"Moshe Berchansky, Omer Levy, Peter Izsak","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-04-15T18:17:12Z","title":"How to Train BERT with an Academic Budget"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2104.07705","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":"sha256:5847ccc4b98e537ea0888accf9cb859cf33a16f3f36a957d32133f11d51a03a2","target":"record","created_at":"2026-07-05T03:12:48Z","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":"80487bc19d6c54c99ffbee20395b4d3358f8448ecb1eff6cdf297ca8f0fa1515","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-04-15T18:17:12Z","title_canon_sha256":"919f1d0ed3526fda30801fa55e08d0efcfc82b0a5248fda1fd476929e2f5bab6"},"schema_version":"1.0","source":{"id":"2104.07705","kind":"arxiv","version":2}},"canonical_sha256":"f6184cd6efdd4a9c2d94a1840b1e828075f166dfc92fb7b820094c49d5f44e69","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f6184cd6efdd4a9c2d94a1840b1e828075f166dfc92fb7b820094c49d5f44e69","first_computed_at":"2026-07-05T03:12:48.112379Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T03:12:48.112379Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"7m3nVYSc75Alen2BspwjAZthAKwwEzVs32q7770IeMz2lLSuAmPl6S16q5wvz3OLBE5s1Qb/DhESRRW2JarcDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T03:12:48.112794Z","signed_message":"canonical_sha256_bytes"},"source_id":"2104.07705","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5847ccc4b98e537ea0888accf9cb859cf33a16f3f36a957d32133f11d51a03a2","sha256:756f7e3b3e11270283deb7f5e3315f49d8f9fa8b7fda80fbeda41810fd0d655a"],"state_sha256":"8c40ab3d842246d17f5a94263a1422bce0cd1bc547203149dfbcc478090ef8b1"}