{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:VFA2TOGHKISBNYULZUQHOQPKW2","short_pith_number":"pith:VFA2TOGH","schema_version":"1.0","canonical_sha256":"a941a9b8c7522416e28bcd207741eab69b5371a44f1218c63a76bf2ccbc622b5","source":{"kind":"arxiv","id":"2110.05367","version":3},"attestation_state":"computed","paper":{"title":"Improving Gender Fairness of Pre-Trained Language Models without Catastrophic Forgetting","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CY","cs.LG"],"primary_cat":"cs.CL","authors_text":"Caiming Xiong, Chen Xing, Wenhao Liu, Zahra Fatemi","submitted_at":"2021-10-11T15:52:16Z","abstract_excerpt":"Existing studies addressing gender bias of pre-trained language models, usually build a small gender-neutral data set and conduct a second phase pre-training on the model with such data. However, given the limited size and concentrated focus of the gender-neutral data, catastrophic forgetting would occur during second-phase pre-training. Forgetting information in the original training data may damage the model's downstream performance by a large margin. In this work, we empirically show that catastrophic forgetting occurs in such methods by evaluating them with general NLP tasks in GLUE. Then,"},"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":"2110.05367","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-10-11T15:52:16Z","cross_cats_sorted":["cs.CY","cs.LG"],"title_canon_sha256":"94623472b75ac1b75683961f3f355b7b8c3cda9e0f95e26a8a783ad32862a9da","abstract_canon_sha256":"4404550360ceacc1ebeab40e4c06fdeef768afa3dd9c690b90e4ed56a2cfb4b2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:26:15.104902Z","signature_b64":"dFhqCo7SEw5WFOY2DEZ3QXHQieYjz3Ua6+O1thnah1sf/Am68CVA5vBlruOa/YGpSQ6+CO8D/RGKhB25jUsvCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a941a9b8c7522416e28bcd207741eab69b5371a44f1218c63a76bf2ccbc622b5","last_reissued_at":"2026-07-05T06:26:15.104337Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:26:15.104337Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improving Gender Fairness of Pre-Trained Language Models without Catastrophic Forgetting","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CY","cs.LG"],"primary_cat":"cs.CL","authors_text":"Caiming Xiong, Chen Xing, Wenhao Liu, Zahra Fatemi","submitted_at":"2021-10-11T15:52:16Z","abstract_excerpt":"Existing studies addressing gender bias of pre-trained language models, usually build a small gender-neutral data set and conduct a second phase pre-training on the model with such data. However, given the limited size and concentrated focus of the gender-neutral data, catastrophic forgetting would occur during second-phase pre-training. Forgetting information in the original training data may damage the model's downstream performance by a large margin. In this work, we empirically show that catastrophic forgetting occurs in such methods by evaluating them with general NLP tasks in GLUE. Then,"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.05367","kind":"arxiv","version":3},"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/2110.05367/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":"2110.05367","created_at":"2026-07-05T06:26:15.104402+00:00"},{"alias_kind":"arxiv_version","alias_value":"2110.05367v3","created_at":"2026-07-05T06:26:15.104402+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.05367","created_at":"2026-07-05T06:26:15.104402+00:00"},{"alias_kind":"pith_short_12","alias_value":"VFA2TOGHKISB","created_at":"2026-07-05T06:26:15.104402+00:00"},{"alias_kind":"pith_short_16","alias_value":"VFA2TOGHKISBNYUL","created_at":"2026-07-05T06:26:15.104402+00:00"},{"alias_kind":"pith_short_8","alias_value":"VFA2TOGH","created_at":"2026-07-05T06:26:15.104402+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2201.11990","citing_title":"Using DeepSpeed and Megatron to Train Megatron-Turing NLG 530B, A Large-Scale Generative Language Model","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VFA2TOGHKISBNYULZUQHOQPKW2","json":"https://pith.science/pith/VFA2TOGHKISBNYULZUQHOQPKW2.json","graph_json":"https://pith.science/api/pith-number/VFA2TOGHKISBNYULZUQHOQPKW2/graph.json","events_json":"https://pith.science/api/pith-number/VFA2TOGHKISBNYULZUQHOQPKW2/events.json","paper":"https://pith.science/paper/VFA2TOGH"},"agent_actions":{"view_html":"https://pith.science/pith/VFA2TOGHKISBNYULZUQHOQPKW2","download_json":"https://pith.science/pith/VFA2TOGHKISBNYULZUQHOQPKW2.json","view_paper":"https://pith.science/paper/VFA2TOGH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2110.05367&json=true","fetch_graph":"https://pith.science/api/pith-number/VFA2TOGHKISBNYULZUQHOQPKW2/graph.json","fetch_events":"https://pith.science/api/pith-number/VFA2TOGHKISBNYULZUQHOQPKW2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VFA2TOGHKISBNYULZUQHOQPKW2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VFA2TOGHKISBNYULZUQHOQPKW2/action/storage_attestation","attest_author":"https://pith.science/pith/VFA2TOGHKISBNYULZUQHOQPKW2/action/author_attestation","sign_citation":"https://pith.science/pith/VFA2TOGHKISBNYULZUQHOQPKW2/action/citation_signature","submit_replication":"https://pith.science/pith/VFA2TOGHKISBNYULZUQHOQPKW2/action/replication_record"}},"created_at":"2026-07-05T06:26:15.104402+00:00","updated_at":"2026-07-05T06:26:15.104402+00:00"}