{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:O4OQRHN7EPJBFWBLDVS66DYIHL","short_pith_number":"pith:O4OQRHN7","schema_version":"1.0","canonical_sha256":"771d089dbf23d212d82b1d65ef0f083afc5c82b52083f9b469672c31138d4ba9","source":{"kind":"arxiv","id":"2210.05497","version":1},"attestation_state":"computed","paper":{"title":"Improving Sharpness-Aware Minimization with Fisher Mask for Better Generalization on Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bo Du, Dacheng Tao, Juhua Liu, Liang Ding, Li Shen, Peng Mi, Qihuang Zhong","submitted_at":"2022-10-11T14:53:58Z","abstract_excerpt":"Fine-tuning large pretrained language models on a limited training corpus usually suffers from poor generalization. Prior works show that the recently-proposed sharpness-aware minimization (SAM) optimization method can improve the model generalization. However, SAM adds a perturbation to each model parameter equally (but not all parameters contribute equally to the optimization of training), which we argue is sub-optimal and will lead to excessive computation. In this paper, we propose a novel optimization procedure, namely FSAM, which introduces a Fisher mask to improve the efficiency and per"},"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":"2210.05497","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-10-11T14:53:58Z","cross_cats_sorted":[],"title_canon_sha256":"d1a1b1b3839365a67b94141564158f2b8f4d8c1c9ac447af91b113515219a8cd","abstract_canon_sha256":"bbc0ae8bac50932764ea53bc1d2017225df8260564b34b31bea6344b96341147"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:05:17.380471Z","signature_b64":"qPi/3+k6aaFXhIkbfzsno4LBmVYPJWLl1nuwBat229WUaYL0sUaCvIO4SDB8cIxiW/dfZWetWySuZVHF8xzaBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"771d089dbf23d212d82b1d65ef0f083afc5c82b52083f9b469672c31138d4ba9","last_reissued_at":"2026-07-05T05:05:17.379974Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:05:17.379974Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improving Sharpness-Aware Minimization with Fisher Mask for Better Generalization on Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bo Du, Dacheng Tao, Juhua Liu, Liang Ding, Li Shen, Peng Mi, Qihuang Zhong","submitted_at":"2022-10-11T14:53:58Z","abstract_excerpt":"Fine-tuning large pretrained language models on a limited training corpus usually suffers from poor generalization. Prior works show that the recently-proposed sharpness-aware minimization (SAM) optimization method can improve the model generalization. However, SAM adds a perturbation to each model parameter equally (but not all parameters contribute equally to the optimization of training), which we argue is sub-optimal and will lead to excessive computation. In this paper, we propose a novel optimization procedure, namely FSAM, which introduces a Fisher mask to improve the efficiency and per"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.05497","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/2210.05497/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":"2210.05497","created_at":"2026-07-05T05:05:17.380032+00:00"},{"alias_kind":"arxiv_version","alias_value":"2210.05497v1","created_at":"2026-07-05T05:05:17.380032+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.05497","created_at":"2026-07-05T05:05:17.380032+00:00"},{"alias_kind":"pith_short_12","alias_value":"O4OQRHN7EPJB","created_at":"2026-07-05T05:05:17.380032+00:00"},{"alias_kind":"pith_short_16","alias_value":"O4OQRHN7EPJBFWBL","created_at":"2026-07-05T05:05:17.380032+00:00"},{"alias_kind":"pith_short_8","alias_value":"O4OQRHN7","created_at":"2026-07-05T05:05:17.380032+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2602.01003","citing_title":"ESSAM: A Novel Competitive Evolution Strategies Approach to Reinforcement Learning for Memory Efficient LLMs Fine-Tuning","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2604.11064","citing_title":"A Faster Path to Continual Learning","ref_index":64,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/O4OQRHN7EPJBFWBLDVS66DYIHL","json":"https://pith.science/pith/O4OQRHN7EPJBFWBLDVS66DYIHL.json","graph_json":"https://pith.science/api/pith-number/O4OQRHN7EPJBFWBLDVS66DYIHL/graph.json","events_json":"https://pith.science/api/pith-number/O4OQRHN7EPJBFWBLDVS66DYIHL/events.json","paper":"https://pith.science/paper/O4OQRHN7"},"agent_actions":{"view_html":"https://pith.science/pith/O4OQRHN7EPJBFWBLDVS66DYIHL","download_json":"https://pith.science/pith/O4OQRHN7EPJBFWBLDVS66DYIHL.json","view_paper":"https://pith.science/paper/O4OQRHN7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2210.05497&json=true","fetch_graph":"https://pith.science/api/pith-number/O4OQRHN7EPJBFWBLDVS66DYIHL/graph.json","fetch_events":"https://pith.science/api/pith-number/O4OQRHN7EPJBFWBLDVS66DYIHL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/O4OQRHN7EPJBFWBLDVS66DYIHL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/O4OQRHN7EPJBFWBLDVS66DYIHL/action/storage_attestation","attest_author":"https://pith.science/pith/O4OQRHN7EPJBFWBLDVS66DYIHL/action/author_attestation","sign_citation":"https://pith.science/pith/O4OQRHN7EPJBFWBLDVS66DYIHL/action/citation_signature","submit_replication":"https://pith.science/pith/O4OQRHN7EPJBFWBLDVS66DYIHL/action/replication_record"}},"created_at":"2026-07-05T05:05:17.380032+00:00","updated_at":"2026-07-05T05:05:17.380032+00:00"}