{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:LDCP5TTIQXBEDIDF7M2HZ4IXRY","short_pith_number":"pith:LDCP5TTI","schema_version":"1.0","canonical_sha256":"58c4fece6885c241a065fb347cf1178e26b778c89d529e7432cef8877dfb7509","source":{"kind":"arxiv","id":"2406.09967","version":1},"attestation_state":"computed","paper":{"title":"Bag of Lies: Robustness in Continuous Pre-training BERT","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ine Gevers, Walter Daelemans","submitted_at":"2024-06-14T12:16:08Z","abstract_excerpt":"This study aims to acquire more insights into the continuous pre-training phase of BERT regarding entity knowledge, using the COVID-19 pandemic as a case study. Since the pandemic emerged after the last update of BERT's pre-training data, the model has little to no entity knowledge about COVID-19. Using continuous pre-training, we control what entity knowledge is available to the model. We compare the baseline BERT model with the further pre-trained variants on the fact-checking benchmark Check-COVID. To test the robustness of continuous pre-training, we experiment with several adversarial met"},"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":"2406.09967","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-14T12:16:08Z","cross_cats_sorted":[],"title_canon_sha256":"be31c76a083c5ca48d27132fa7a0deaf66b047756b2dd21aa013eca09b3f9d15","abstract_canon_sha256":"9ad6ce3f1e3f44182fa1e1566c1053978c07d6180f23b11596a9933c1aed4384"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:32:02.529695Z","signature_b64":"MQHVmEwfqU46ohSnNwrOj/sGBJTeX/dQEtnLi5cRh7uhBz1Hlb9ABiqy8gaN3tz5CWn1th/N3iBbgra8KFLqDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"58c4fece6885c241a065fb347cf1178e26b778c89d529e7432cef8877dfb7509","last_reissued_at":"2026-07-05T08:32:02.529204Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:32:02.529204Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Bag of Lies: Robustness in Continuous Pre-training BERT","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ine Gevers, Walter Daelemans","submitted_at":"2024-06-14T12:16:08Z","abstract_excerpt":"This study aims to acquire more insights into the continuous pre-training phase of BERT regarding entity knowledge, using the COVID-19 pandemic as a case study. Since the pandemic emerged after the last update of BERT's pre-training data, the model has little to no entity knowledge about COVID-19. Using continuous pre-training, we control what entity knowledge is available to the model. We compare the baseline BERT model with the further pre-trained variants on the fact-checking benchmark Check-COVID. To test the robustness of continuous pre-training, we experiment with several adversarial met"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.09967","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/2406.09967/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":"2406.09967","created_at":"2026-07-05T08:32:02.529254+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.09967v1","created_at":"2026-07-05T08:32:02.529254+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.09967","created_at":"2026-07-05T08:32:02.529254+00:00"},{"alias_kind":"pith_short_12","alias_value":"LDCP5TTIQXBE","created_at":"2026-07-05T08:32:02.529254+00:00"},{"alias_kind":"pith_short_16","alias_value":"LDCP5TTIQXBEDIDF","created_at":"2026-07-05T08:32:02.529254+00:00"},{"alias_kind":"pith_short_8","alias_value":"LDCP5TTI","created_at":"2026-07-05T08:32:02.529254+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2608.04034","citing_title":"A Multimodal Automatic Redteaming Evaluation based on Atomic Jailbreak Strategy Decoupling and Combination","ref_index":51,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LDCP5TTIQXBEDIDF7M2HZ4IXRY","json":"https://pith.science/pith/LDCP5TTIQXBEDIDF7M2HZ4IXRY.json","graph_json":"https://pith.science/api/pith-number/LDCP5TTIQXBEDIDF7M2HZ4IXRY/graph.json","events_json":"https://pith.science/api/pith-number/LDCP5TTIQXBEDIDF7M2HZ4IXRY/events.json","paper":"https://pith.science/paper/LDCP5TTI"},"agent_actions":{"view_html":"https://pith.science/pith/LDCP5TTIQXBEDIDF7M2HZ4IXRY","download_json":"https://pith.science/pith/LDCP5TTIQXBEDIDF7M2HZ4IXRY.json","view_paper":"https://pith.science/paper/LDCP5TTI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.09967&json=true","fetch_graph":"https://pith.science/api/pith-number/LDCP5TTIQXBEDIDF7M2HZ4IXRY/graph.json","fetch_events":"https://pith.science/api/pith-number/LDCP5TTIQXBEDIDF7M2HZ4IXRY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LDCP5TTIQXBEDIDF7M2HZ4IXRY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LDCP5TTIQXBEDIDF7M2HZ4IXRY/action/storage_attestation","attest_author":"https://pith.science/pith/LDCP5TTIQXBEDIDF7M2HZ4IXRY/action/author_attestation","sign_citation":"https://pith.science/pith/LDCP5TTIQXBEDIDF7M2HZ4IXRY/action/citation_signature","submit_replication":"https://pith.science/pith/LDCP5TTIQXBEDIDF7M2HZ4IXRY/action/replication_record"}},"created_at":"2026-07-05T08:32:02.529254+00:00","updated_at":"2026-07-05T08:32:02.529254+00:00"}