{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:TFOTF73YMOHIII5HGC3QNIU7WQ","short_pith_number":"pith:TFOTF73Y","schema_version":"1.0","canonical_sha256":"995d32ff78638e8423a730b706a29fb41e3d56c158a5526dc875e8d94539cedf","source":{"kind":"arxiv","id":"2312.10136","version":3},"attestation_state":"computed","paper":{"title":"Gradient-based Parameter Selection for Efficient Fine-Tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Ekaterina Shutova, Qizhe Zhang, Renrui Zhang, Shanghang Zhang, Shiji Zhou, Zhi Zhang, Zijun Gao","submitted_at":"2023-12-15T18:59:05Z","abstract_excerpt":"With the growing size of pre-trained models, full fine-tuning and storing all the parameters for various downstream tasks is costly and infeasible. In this paper, we propose a new parameter-efficient fine-tuning method, Gradient-based Parameter Selection (GPS), demonstrating that only tuning a few selected parameters from the pre-trained model while keeping the remainder of the model frozen can generate similar or better performance compared with the full model fine-tuning method. Different from the existing popular and state-of-the-art parameter-efficient fine-tuning approaches, our method do"},"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":"2312.10136","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-12-15T18:59:05Z","cross_cats_sorted":[],"title_canon_sha256":"51807a5066b4e2f78a6e03a9f535f682e79113be3f35affcb61e376b50269155","abstract_canon_sha256":"291d48496e7f3e96fb7ca35e59199c8fc381b855bda1bac59b7e7505704d8d2a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:41:30.352152Z","signature_b64":"olzVFbON4Q4xZ5lipmy6FJr+uYAyS46751vAV1vEodZg2e+V0zK7zXvjjjatErhFPpovnBjbC2I5jp8qMmaSDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"995d32ff78638e8423a730b706a29fb41e3d56c158a5526dc875e8d94539cedf","last_reissued_at":"2026-07-05T09:41:30.351597Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:41:30.351597Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Gradient-based Parameter Selection for Efficient Fine-Tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Ekaterina Shutova, Qizhe Zhang, Renrui Zhang, Shanghang Zhang, Shiji Zhou, Zhi Zhang, Zijun Gao","submitted_at":"2023-12-15T18:59:05Z","abstract_excerpt":"With the growing size of pre-trained models, full fine-tuning and storing all the parameters for various downstream tasks is costly and infeasible. In this paper, we propose a new parameter-efficient fine-tuning method, Gradient-based Parameter Selection (GPS), demonstrating that only tuning a few selected parameters from the pre-trained model while keeping the remainder of the model frozen can generate similar or better performance compared with the full model fine-tuning method. Different from the existing popular and state-of-the-art parameter-efficient fine-tuning approaches, our method do"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.10136","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/2312.10136/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":"2312.10136","created_at":"2026-07-05T09:41:30.351671+00:00"},{"alias_kind":"arxiv_version","alias_value":"2312.10136v3","created_at":"2026-07-05T09:41:30.351671+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.10136","created_at":"2026-07-05T09:41:30.351671+00:00"},{"alias_kind":"pith_short_12","alias_value":"TFOTF73YMOHI","created_at":"2026-07-05T09:41:30.351671+00:00"},{"alias_kind":"pith_short_16","alias_value":"TFOTF73YMOHIII5H","created_at":"2026-07-05T09:41:30.351671+00:00"},{"alias_kind":"pith_short_8","alias_value":"TFOTF73Y","created_at":"2026-07-05T09:41:30.351671+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.16191","citing_title":"GEM: A Scale-Aware and Distribution-Sensitive Sparse Fine-Tuning Framework for Effective Downstream Adaptation","ref_index":45,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TFOTF73YMOHIII5HGC3QNIU7WQ","json":"https://pith.science/pith/TFOTF73YMOHIII5HGC3QNIU7WQ.json","graph_json":"https://pith.science/api/pith-number/TFOTF73YMOHIII5HGC3QNIU7WQ/graph.json","events_json":"https://pith.science/api/pith-number/TFOTF73YMOHIII5HGC3QNIU7WQ/events.json","paper":"https://pith.science/paper/TFOTF73Y"},"agent_actions":{"view_html":"https://pith.science/pith/TFOTF73YMOHIII5HGC3QNIU7WQ","download_json":"https://pith.science/pith/TFOTF73YMOHIII5HGC3QNIU7WQ.json","view_paper":"https://pith.science/paper/TFOTF73Y","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2312.10136&json=true","fetch_graph":"https://pith.science/api/pith-number/TFOTF73YMOHIII5HGC3QNIU7WQ/graph.json","fetch_events":"https://pith.science/api/pith-number/TFOTF73YMOHIII5HGC3QNIU7WQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TFOTF73YMOHIII5HGC3QNIU7WQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TFOTF73YMOHIII5HGC3QNIU7WQ/action/storage_attestation","attest_author":"https://pith.science/pith/TFOTF73YMOHIII5HGC3QNIU7WQ/action/author_attestation","sign_citation":"https://pith.science/pith/TFOTF73YMOHIII5HGC3QNIU7WQ/action/citation_signature","submit_replication":"https://pith.science/pith/TFOTF73YMOHIII5HGC3QNIU7WQ/action/replication_record"}},"created_at":"2026-07-05T09:41:30.351671+00:00","updated_at":"2026-07-05T09:41:30.351671+00:00"}