{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:IU2EGTIRNE4MCPHC5K6J6VS7QT","short_pith_number":"pith:IU2EGTIR","schema_version":"1.0","canonical_sha256":"4534434d116938c13ce2eabc9f565f84ddb3f23c906018844dd98e6d6d77b0ae","source":{"kind":"arxiv","id":"2110.07143","version":1},"attestation_state":"computed","paper":{"title":"bert2BERT: Towards Reusable Pretrained Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Cheng Chen, Fengyu Wang, Lifeng Shang, Qun Liu, Xiao Chen, Xin Jiang, Yichun Yin, Yujia Qin, Zhi Wang, Zhiyuan Liu","submitted_at":"2021-10-14T04:05:25Z","abstract_excerpt":"In recent years, researchers tend to pre-train ever-larger language models to explore the upper limit of deep models. However, large language model pre-training costs intensive computational resources and most of the models are trained from scratch without reusing the existing pre-trained models, which is wasteful. In this paper, we propose bert2BERT, which can effectively transfer the knowledge of an existing smaller pre-trained model (e.g., BERT_BASE) to a large model (e.g., BERT_LARGE) through parameter initialization and significantly improve the pre-training efficiency of the large model."},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2110.07143","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-10-14T04:05:25Z","cross_cats_sorted":[],"title_canon_sha256":"e25e6e21469b33e82b5109bdae0ebb0372eced5f236ed10c4614b642511f9fd3","abstract_canon_sha256":"131a39207ffe16878f620a32f92f87282d7f99bf4b83ae217e4a396a70524d8f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:22:39.317249Z","signature_b64":"/P2ZvgwxEpGmHdXCVB8LzvVc8b9dQyXQP9+ansju1bjMdEYDD1CYxBIn+c7brG9bxIVEKlkTf1IlgmVNJUm+Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4534434d116938c13ce2eabc9f565f84ddb3f23c906018844dd98e6d6d77b0ae","last_reissued_at":"2026-07-05T03:22:39.316815Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:22:39.316815Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"bert2BERT: Towards Reusable Pretrained Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Cheng Chen, Fengyu Wang, Lifeng Shang, Qun Liu, Xiao Chen, Xin Jiang, Yichun Yin, Yujia Qin, Zhi Wang, Zhiyuan Liu","submitted_at":"2021-10-14T04:05:25Z","abstract_excerpt":"In recent years, researchers tend to pre-train ever-larger language models to explore the upper limit of deep models. However, large language model pre-training costs intensive computational resources and most of the models are trained from scratch without reusing the existing pre-trained models, which is wasteful. In this paper, we propose bert2BERT, which can effectively transfer the knowledge of an existing smaller pre-trained model (e.g., BERT_BASE) to a large model (e.g., BERT_LARGE) through parameter initialization and significantly improve the pre-training efficiency of the large model."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.07143","kind":"arxiv","version":1},"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/2110.07143/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2110.07143","created_at":"2026-07-05T03:22:39.316867+00:00"},{"alias_kind":"arxiv_version","alias_value":"2110.07143v1","created_at":"2026-07-05T03:22:39.316867+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.07143","created_at":"2026-07-05T03:22:39.316867+00:00"},{"alias_kind":"pith_short_12","alias_value":"IU2EGTIRNE4M","created_at":"2026-07-05T03:22:39.316867+00:00"},{"alias_kind":"pith_short_16","alias_value":"IU2EGTIRNE4MCPHC","created_at":"2026-07-05T03:22:39.316867+00:00"},{"alias_kind":"pith_short_8","alias_value":"IU2EGTIR","created_at":"2026-07-05T03:22:39.316867+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.11953","citing_title":"IAM: Efficient Inference through Attention Mapping between Different-scale LLMs","ref_index":3,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IU2EGTIRNE4MCPHC5K6J6VS7QT","json":"https://pith.science/pith/IU2EGTIRNE4MCPHC5K6J6VS7QT.json","graph_json":"https://pith.science/api/pith-number/IU2EGTIRNE4MCPHC5K6J6VS7QT/graph.json","events_json":"https://pith.science/api/pith-number/IU2EGTIRNE4MCPHC5K6J6VS7QT/events.json","paper":"https://pith.science/paper/IU2EGTIR"},"agent_actions":{"view_html":"https://pith.science/pith/IU2EGTIRNE4MCPHC5K6J6VS7QT","download_json":"https://pith.science/pith/IU2EGTIRNE4MCPHC5K6J6VS7QT.json","view_paper":"https://pith.science/paper/IU2EGTIR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2110.07143&json=true","fetch_graph":"https://pith.science/api/pith-number/IU2EGTIRNE4MCPHC5K6J6VS7QT/graph.json","fetch_events":"https://pith.science/api/pith-number/IU2EGTIRNE4MCPHC5K6J6VS7QT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IU2EGTIRNE4MCPHC5K6J6VS7QT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IU2EGTIRNE4MCPHC5K6J6VS7QT/action/storage_attestation","attest_author":"https://pith.science/pith/IU2EGTIRNE4MCPHC5K6J6VS7QT/action/author_attestation","sign_citation":"https://pith.science/pith/IU2EGTIRNE4MCPHC5K6J6VS7QT/action/citation_signature","submit_replication":"https://pith.science/pith/IU2EGTIRNE4MCPHC5K6J6VS7QT/action/replication_record"}},"created_at":"2026-07-05T03:22:39.316867+00:00","updated_at":"2026-07-05T03:22:39.316867+00:00"}