{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:QA7RWNZVHSOZY545FIPHJ75CLY","short_pith_number":"pith:QA7RWNZV","canonical_record":{"source":{"id":"2312.17482","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-12-29T06:05:19Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"2ec01f9113ae79dbfac3c116168bf04a50e9bc8c0c914570be20751c6f74186a","abstract_canon_sha256":"1aaa928e67965c78bb1ef340cd75f556ffc0571e9cbeba93ee344781902a6b2a"},"schema_version":"1.0"},"canonical_sha256":"803f1b37353c9d9c779d2a1e74ffa25e3002b4e6bc22e9ca354112bd77c2f9b7","source":{"kind":"arxiv","id":"2312.17482","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2312.17482","created_at":"2026-07-05T07:33:59Z"},{"alias_kind":"arxiv_version","alias_value":"2312.17482v2","created_at":"2026-07-05T07:33:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.17482","created_at":"2026-07-05T07:33:59Z"},{"alias_kind":"pith_short_12","alias_value":"QA7RWNZVHSOZ","created_at":"2026-07-05T07:33:59Z"},{"alias_kind":"pith_short_16","alias_value":"QA7RWNZVHSOZY545","created_at":"2026-07-05T07:33:59Z"},{"alias_kind":"pith_short_8","alias_value":"QA7RWNZV","created_at":"2026-07-05T07:33:59Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:QA7RWNZVHSOZY545FIPHJ75CLY","target":"record","payload":{"canonical_record":{"source":{"id":"2312.17482","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-12-29T06:05:19Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"2ec01f9113ae79dbfac3c116168bf04a50e9bc8c0c914570be20751c6f74186a","abstract_canon_sha256":"1aaa928e67965c78bb1ef340cd75f556ffc0571e9cbeba93ee344781902a6b2a"},"schema_version":"1.0"},"canonical_sha256":"803f1b37353c9d9c779d2a1e74ffa25e3002b4e6bc22e9ca354112bd77c2f9b7","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:33:59.708082Z","signature_b64":"YRz0d5r23utTLAiObLH84swD6h9I0L1SL7twpZYhJ1fNXLNOIk87cXZ9gTQTokaTBz4jmHoPVoqGNW32yzt8Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"803f1b37353c9d9c779d2a1e74ffa25e3002b4e6bc22e9ca354112bd77c2f9b7","last_reissued_at":"2026-07-05T07:33:59.707683Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:33:59.707683Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2312.17482","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":"2026-07-05T07:33:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2ExIDnRX71wywxz4FdbgnaQY2ZpEABfjoNqKXZ3Pk2vLHkw8SfweuwxOeTKHbmaIWAgKhF2103pW1MikFQ9gBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T17:44:50.270817Z"},"content_sha256":"e3b5180edca2d0834b038ddf7588e09d88892093fc7a4b425ceae03cde4e5e64","schema_version":"1.0","event_id":"sha256:e3b5180edca2d0834b038ddf7588e09d88892093fc7a4b425ceae03cde4e5e64"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:QA7RWNZVHSOZY545FIPHJ75CLY","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"MosaicBERT: A Bidirectional Encoder Optimized for Fast Pretraining","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Abhinav Venigalla, Alex Trott, Daniel King, Daya Khudia, Jacob Portes, Jonathan Frankle, Moin Nadeem, Nikhil Sardana, Sam Havens","submitted_at":"2023-12-29T06:05:19Z","abstract_excerpt":"Although BERT-style encoder models are heavily used in NLP research, many researchers do not pretrain their own BERTs from scratch due to the high cost of training. In the past half-decade since BERT first rose to prominence, many advances have been made with other transformer architectures and training configurations that have yet to be systematically incorporated into BERT. Here, we introduce MosaicBERT, a BERT-style encoder architecture and training recipe that is empirically optimized for fast pretraining. This efficient architecture incorporates FlashAttention, Attention with Linear Biase"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.17482","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/2312.17482/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":"2026-07-05T07:33:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Id0Q5fKj4CNQuvLBkqRDUG6hUc+7HfOMjKKFcv4BcCRjBPYet0M3ODp6SsUmxdELMwCAnPPA8LWdXmXaTzteBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T17:44:50.271192Z"},"content_sha256":"42472c28f3c9a20e9c18023a89fba3e6f75c47538ea5038f915e9824c3836d99","schema_version":"1.0","event_id":"sha256:42472c28f3c9a20e9c18023a89fba3e6f75c47538ea5038f915e9824c3836d99"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QA7RWNZVHSOZY545FIPHJ75CLY/bundle.json","state_url":"https://pith.science/pith/QA7RWNZVHSOZY545FIPHJ75CLY/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QA7RWNZVHSOZY545FIPHJ75CLY/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":"2026-07-06T17:44:50Z","links":{"resolver":"https://pith.science/pith/QA7RWNZVHSOZY545FIPHJ75CLY","bundle":"https://pith.science/pith/QA7RWNZVHSOZY545FIPHJ75CLY/bundle.json","state":"https://pith.science/pith/QA7RWNZVHSOZY545FIPHJ75CLY/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QA7RWNZVHSOZY545FIPHJ75CLY/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:QA7RWNZVHSOZY545FIPHJ75CLY","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":"1aaa928e67965c78bb1ef340cd75f556ffc0571e9cbeba93ee344781902a6b2a","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-12-29T06:05:19Z","title_canon_sha256":"2ec01f9113ae79dbfac3c116168bf04a50e9bc8c0c914570be20751c6f74186a"},"schema_version":"1.0","source":{"id":"2312.17482","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2312.17482","created_at":"2026-07-05T07:33:59Z"},{"alias_kind":"arxiv_version","alias_value":"2312.17482v2","created_at":"2026-07-05T07:33:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.17482","created_at":"2026-07-05T07:33:59Z"},{"alias_kind":"pith_short_12","alias_value":"QA7RWNZVHSOZ","created_at":"2026-07-05T07:33:59Z"},{"alias_kind":"pith_short_16","alias_value":"QA7RWNZVHSOZY545","created_at":"2026-07-05T07:33:59Z"},{"alias_kind":"pith_short_8","alias_value":"QA7RWNZV","created_at":"2026-07-05T07:33:59Z"}],"graph_snapshots":[{"event_id":"sha256:42472c28f3c9a20e9c18023a89fba3e6f75c47538ea5038f915e9824c3836d99","target":"graph","created_at":"2026-07-05T07:33:59Z","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/2312.17482/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Although BERT-style encoder models are heavily used in NLP research, many researchers do not pretrain their own BERTs from scratch due to the high cost of training. In the past half-decade since BERT first rose to prominence, many advances have been made with other transformer architectures and training configurations that have yet to be systematically incorporated into BERT. Here, we introduce MosaicBERT, a BERT-style encoder architecture and training recipe that is empirically optimized for fast pretraining. This efficient architecture incorporates FlashAttention, Attention with Linear Biase","authors_text":"Abhinav Venigalla, Alex Trott, Daniel King, Daya Khudia, Jacob Portes, Jonathan Frankle, Moin Nadeem, Nikhil Sardana, Sam Havens","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-12-29T06:05:19Z","title":"MosaicBERT: A Bidirectional Encoder Optimized for Fast Pretraining"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.17482","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:e3b5180edca2d0834b038ddf7588e09d88892093fc7a4b425ceae03cde4e5e64","target":"record","created_at":"2026-07-05T07:33:59Z","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":"1aaa928e67965c78bb1ef340cd75f556ffc0571e9cbeba93ee344781902a6b2a","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-12-29T06:05:19Z","title_canon_sha256":"2ec01f9113ae79dbfac3c116168bf04a50e9bc8c0c914570be20751c6f74186a"},"schema_version":"1.0","source":{"id":"2312.17482","kind":"arxiv","version":2}},"canonical_sha256":"803f1b37353c9d9c779d2a1e74ffa25e3002b4e6bc22e9ca354112bd77c2f9b7","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"803f1b37353c9d9c779d2a1e74ffa25e3002b4e6bc22e9ca354112bd77c2f9b7","first_computed_at":"2026-07-05T07:33:59.707683Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:33:59.707683Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"YRz0d5r23utTLAiObLH84swD6h9I0L1SL7twpZYhJ1fNXLNOIk87cXZ9gTQTokaTBz4jmHoPVoqGNW32yzt8Aw==","signature_status":"signed_v1","signed_at":"2026-07-05T07:33:59.708082Z","signed_message":"canonical_sha256_bytes"},"source_id":"2312.17482","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e3b5180edca2d0834b038ddf7588e09d88892093fc7a4b425ceae03cde4e5e64","sha256:42472c28f3c9a20e9c18023a89fba3e6f75c47538ea5038f915e9824c3836d99"],"state_sha256":"5b88e5033f351dfe10cc7d378fbd03fdb4edba4dfb6b4410ebc9106878199071"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qhsW2fT+Aj8sZUNrDmzrv9mJL8KPKeOzJDmJZTSYNoGYcDeJqKgmp2YKm01rv3A/2Ia1JZjgVCP4yJgSMb5wBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-06T17:44:50.273111Z","bundle_sha256":"e59b96dbda6018ce1db105e25dc914bfa9ea6ed3e513c0892baeb1fbfd5cff28"}}