{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:BPRS45ARI2VCHIDLXI66KTIF3W","short_pith_number":"pith:BPRS45AR","schema_version":"1.0","canonical_sha256":"0be32e741146aa23a06bba3de54d05dd92a62ca6ba79af135a18f94eb2eb0b99","source":{"kind":"arxiv","id":"2107.07445","version":2},"attestation_state":"computed","paper":{"title":"AutoBERT-Zero: Evolving BERT Backbone from Scratch","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Hang Xu, Han Shi, Jiahui Gao, Philip L.H. Yu, Xiaodan Liang, Xiaozhe Ren, Xin Jiang, Zhenguo Li","submitted_at":"2021-07-15T16:46:01Z","abstract_excerpt":"Transformer-based pre-trained language models like BERT and its variants have recently achieved promising performance in various natural language processing (NLP) tasks. However, the conventional paradigm constructs the backbone by purely stacking the manually designed global self-attention layers, introducing inductive bias and thus leads to sub-optimal. In this work, we make the first attempt to automatically discover novel pre-trained language model (PLM) backbone on a flexible search space containing the most fundamental operations from scratch. Specifically, we propose a well-designed sea"},"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":"2107.07445","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-07-15T16:46:01Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"e41ac6fcde0c0e71a3da2f148c2d1f1a313b7152809967a78cf7be5faa70c580","abstract_canon_sha256":"a64cb290b1821adb6c9db8c272728bd347d88ebf56c1831542e644c5d140e929"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:54:22.172899Z","signature_b64":"lXBPsgyCPr0m15sxPELFbHqI9Ng7GNxTCb9r3zEcOz/KdWdlTtFnNfO5RJgIK/o2JAur1JDQQX26bTAadoDrCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0be32e741146aa23a06bba3de54d05dd92a62ca6ba79af135a18f94eb2eb0b99","last_reissued_at":"2026-07-05T03:54:22.172498Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:54:22.172498Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"AutoBERT-Zero: Evolving BERT Backbone from Scratch","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Hang Xu, Han Shi, Jiahui Gao, Philip L.H. Yu, Xiaodan Liang, Xiaozhe Ren, Xin Jiang, Zhenguo Li","submitted_at":"2021-07-15T16:46:01Z","abstract_excerpt":"Transformer-based pre-trained language models like BERT and its variants have recently achieved promising performance in various natural language processing (NLP) tasks. However, the conventional paradigm constructs the backbone by purely stacking the manually designed global self-attention layers, introducing inductive bias and thus leads to sub-optimal. In this work, we make the first attempt to automatically discover novel pre-trained language model (PLM) backbone on a flexible search space containing the most fundamental operations from scratch. Specifically, we propose a well-designed sea"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.07445","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/2107.07445/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":"2107.07445","created_at":"2026-07-05T03:54:22.172551+00:00"},{"alias_kind":"arxiv_version","alias_value":"2107.07445v2","created_at":"2026-07-05T03:54:22.172551+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.07445","created_at":"2026-07-05T03:54:22.172551+00:00"},{"alias_kind":"pith_short_12","alias_value":"BPRS45ARI2VC","created_at":"2026-07-05T03:54:22.172551+00:00"},{"alias_kind":"pith_short_16","alias_value":"BPRS45ARI2VCHIDL","created_at":"2026-07-05T03:54:22.172551+00:00"},{"alias_kind":"pith_short_8","alias_value":"BPRS45AR","created_at":"2026-07-05T03:54:22.172551+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BPRS45ARI2VCHIDLXI66KTIF3W","json":"https://pith.science/pith/BPRS45ARI2VCHIDLXI66KTIF3W.json","graph_json":"https://pith.science/api/pith-number/BPRS45ARI2VCHIDLXI66KTIF3W/graph.json","events_json":"https://pith.science/api/pith-number/BPRS45ARI2VCHIDLXI66KTIF3W/events.json","paper":"https://pith.science/paper/BPRS45AR"},"agent_actions":{"view_html":"https://pith.science/pith/BPRS45ARI2VCHIDLXI66KTIF3W","download_json":"https://pith.science/pith/BPRS45ARI2VCHIDLXI66KTIF3W.json","view_paper":"https://pith.science/paper/BPRS45AR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2107.07445&json=true","fetch_graph":"https://pith.science/api/pith-number/BPRS45ARI2VCHIDLXI66KTIF3W/graph.json","fetch_events":"https://pith.science/api/pith-number/BPRS45ARI2VCHIDLXI66KTIF3W/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BPRS45ARI2VCHIDLXI66KTIF3W/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BPRS45ARI2VCHIDLXI66KTIF3W/action/storage_attestation","attest_author":"https://pith.science/pith/BPRS45ARI2VCHIDLXI66KTIF3W/action/author_attestation","sign_citation":"https://pith.science/pith/BPRS45ARI2VCHIDLXI66KTIF3W/action/citation_signature","submit_replication":"https://pith.science/pith/BPRS45ARI2VCHIDLXI66KTIF3W/action/replication_record"}},"created_at":"2026-07-05T03:54:22.172551+00:00","updated_at":"2026-07-05T03:54:22.172551+00:00"}