{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:J4RIYE5R7CG3LHZ4GDKOFGGEK5","short_pith_number":"pith:J4RIYE5R","schema_version":"1.0","canonical_sha256":"4f228c13b1f88db59f3c30d4e298c4574280e379c388aa8de5afc9be08142063","source":{"kind":"arxiv","id":"2402.12220","version":3},"attestation_state":"computed","paper":{"title":"Bayesian Parameter-Efficient Fine-Tuning for Overcoming Catastrophic Forgetting","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"eess.AS","authors_text":"Haolin Chen, Philip N. Garner","submitted_at":"2024-02-19T15:26:19Z","abstract_excerpt":"We are motivated primarily by the adaptation of text-to-speech synthesis models; however we argue that more generic parameter-efficient fine-tuning (PEFT) is an appropriate framework to do such adaptation. Nevertheless, catastrophic forgetting remains an issue with PEFT, damaging the pre-trained model's inherent capabilities. We demonstrate that existing Bayesian learning techniques can be applied to PEFT to prevent catastrophic forgetting as long as the parameter shift of the fine-tuned layers can be calculated differentiably. In a principled series of experiments on language modeling and spe"},"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":"2402.12220","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.AS","submitted_at":"2024-02-19T15:26:19Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"5fd1cc3ff5ca1f353be8edbf04fb7dee2cc41693c94b7460498744b09309a45a","abstract_canon_sha256":"5d8d72935f72f61ad072f5ccf331df545c58524c00a9bf7bd4ed799d37ccb466"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:45:44.642489Z","signature_b64":"8HLMqMRn9yILDmra6qGoM88jQ9LZ1UfzGjFZP0udjOG7k7a0N6HheeE7tWfp0TQG0sypDtgABIU8UXL0ZiZeCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4f228c13b1f88db59f3c30d4e298c4574280e379c388aa8de5afc9be08142063","last_reissued_at":"2026-07-05T09:45:44.641978Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:45:44.641978Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Bayesian Parameter-Efficient Fine-Tuning for Overcoming Catastrophic Forgetting","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"eess.AS","authors_text":"Haolin Chen, Philip N. Garner","submitted_at":"2024-02-19T15:26:19Z","abstract_excerpt":"We are motivated primarily by the adaptation of text-to-speech synthesis models; however we argue that more generic parameter-efficient fine-tuning (PEFT) is an appropriate framework to do such adaptation. Nevertheless, catastrophic forgetting remains an issue with PEFT, damaging the pre-trained model's inherent capabilities. We demonstrate that existing Bayesian learning techniques can be applied to PEFT to prevent catastrophic forgetting as long as the parameter shift of the fine-tuned layers can be calculated differentiably. In a principled series of experiments on language modeling and spe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.12220","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/2402.12220/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":"2402.12220","created_at":"2026-07-05T09:45:44.642037+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.12220v3","created_at":"2026-07-05T09:45:44.642037+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.12220","created_at":"2026-07-05T09:45:44.642037+00:00"},{"alias_kind":"pith_short_12","alias_value":"J4RIYE5R7CG3","created_at":"2026-07-05T09:45:44.642037+00:00"},{"alias_kind":"pith_short_16","alias_value":"J4RIYE5R7CG3LHZ4","created_at":"2026-07-05T09:45:44.642037+00:00"},{"alias_kind":"pith_short_8","alias_value":"J4RIYE5R","created_at":"2026-07-05T09:45:44.642037+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/J4RIYE5R7CG3LHZ4GDKOFGGEK5","json":"https://pith.science/pith/J4RIYE5R7CG3LHZ4GDKOFGGEK5.json","graph_json":"https://pith.science/api/pith-number/J4RIYE5R7CG3LHZ4GDKOFGGEK5/graph.json","events_json":"https://pith.science/api/pith-number/J4RIYE5R7CG3LHZ4GDKOFGGEK5/events.json","paper":"https://pith.science/paper/J4RIYE5R"},"agent_actions":{"view_html":"https://pith.science/pith/J4RIYE5R7CG3LHZ4GDKOFGGEK5","download_json":"https://pith.science/pith/J4RIYE5R7CG3LHZ4GDKOFGGEK5.json","view_paper":"https://pith.science/paper/J4RIYE5R","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.12220&json=true","fetch_graph":"https://pith.science/api/pith-number/J4RIYE5R7CG3LHZ4GDKOFGGEK5/graph.json","fetch_events":"https://pith.science/api/pith-number/J4RIYE5R7CG3LHZ4GDKOFGGEK5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/J4RIYE5R7CG3LHZ4GDKOFGGEK5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/J4RIYE5R7CG3LHZ4GDKOFGGEK5/action/storage_attestation","attest_author":"https://pith.science/pith/J4RIYE5R7CG3LHZ4GDKOFGGEK5/action/author_attestation","sign_citation":"https://pith.science/pith/J4RIYE5R7CG3LHZ4GDKOFGGEK5/action/citation_signature","submit_replication":"https://pith.science/pith/J4RIYE5R7CG3LHZ4GDKOFGGEK5/action/replication_record"}},"created_at":"2026-07-05T09:45:44.642037+00:00","updated_at":"2026-07-05T09:45:44.642037+00:00"}