{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:YWGZE6AI4C2FDBL7RFIK2M5AU7","short_pith_number":"pith:YWGZE6AI","schema_version":"1.0","canonical_sha256":"c58d927808e0b451857f8950ad33a0a7cb4ca0ac55572d993f71577f3623be88","source":{"kind":"arxiv","id":"2510.00419","version":2},"attestation_state":"computed","paper":{"title":"Learning a Zeroth-Order Optimizer for Fine-Tuning LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Haoyu Li, Huan Zhang, Kairun Zhang, Yanjun Zhao, Yifan Sun","submitted_at":"2025-10-01T02:01:07Z","abstract_excerpt":"Zeroth-order optimizers have recently emerged as an attractive approach for fine-tuning large language models (LLMs), as they avoid backpropagation and can substantially reduce memory overhead relative to standard first-order training. However, existing zeroth-order methods rely on hand-crafted, static sampling strategies that are not adaptable to model-specific structures. To address this, we propose ZO-Finetuner, a learning-based zeroth-order optimizer for LLMs that automatically learns efficient perturbation strategies through a compact and memory-efficient design. Motivated by the fact tha"},"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":"2510.00419","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-10-01T02:01:07Z","cross_cats_sorted":[],"title_canon_sha256":"cadfa8e3346c537d8bbc03384585c3e7d74fc88041f1111e4cf36c74ffa44ca5","abstract_canon_sha256":"ffb3948a508989e8ede5702657ee4b4027e8b8fcced5153ffcac1f7ad7ea97f9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-01T01:02:22.073852Z","signature_b64":"kQpSt59cIxgjhVWTUm+NoS30mWK7XRHqkrQaFP0kpjDgYDOt3lMGbJU4+StQ/POxqiNO2Du7PJgyHPNTsew0DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c58d927808e0b451857f8950ad33a0a7cb4ca0ac55572d993f71577f3623be88","last_reissued_at":"2026-06-01T01:02:22.072909Z","signature_status":"signed_v1","first_computed_at":"2026-06-01T01:02:22.072909Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning a Zeroth-Order Optimizer for Fine-Tuning LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Haoyu Li, Huan Zhang, Kairun Zhang, Yanjun Zhao, Yifan Sun","submitted_at":"2025-10-01T02:01:07Z","abstract_excerpt":"Zeroth-order optimizers have recently emerged as an attractive approach for fine-tuning large language models (LLMs), as they avoid backpropagation and can substantially reduce memory overhead relative to standard first-order training. However, existing zeroth-order methods rely on hand-crafted, static sampling strategies that are not adaptable to model-specific structures. To address this, we propose ZO-Finetuner, a learning-based zeroth-order optimizer for LLMs that automatically learns efficient perturbation strategies through a compact and memory-efficient design. Motivated by the fact tha"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2510.00419","kind":"arxiv","version":2},"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/2510.00419/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":"2510.00419","created_at":"2026-06-01T01:02:22.073023+00:00"},{"alias_kind":"arxiv_version","alias_value":"2510.00419v2","created_at":"2026-06-01T01:02:22.073023+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2510.00419","created_at":"2026-06-01T01:02:22.073023+00:00"},{"alias_kind":"pith_short_12","alias_value":"YWGZE6AI4C2F","created_at":"2026-06-01T01:02:22.073023+00:00"},{"alias_kind":"pith_short_16","alias_value":"YWGZE6AI4C2FDBL7","created_at":"2026-06-01T01:02:22.073023+00:00"},{"alias_kind":"pith_short_8","alias_value":"YWGZE6AI","created_at":"2026-06-01T01:02:22.073023+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/YWGZE6AI4C2FDBL7RFIK2M5AU7","json":"https://pith.science/pith/YWGZE6AI4C2FDBL7RFIK2M5AU7.json","graph_json":"https://pith.science/api/pith-number/YWGZE6AI4C2FDBL7RFIK2M5AU7/graph.json","events_json":"https://pith.science/api/pith-number/YWGZE6AI4C2FDBL7RFIK2M5AU7/events.json","paper":"https://pith.science/paper/YWGZE6AI"},"agent_actions":{"view_html":"https://pith.science/pith/YWGZE6AI4C2FDBL7RFIK2M5AU7","download_json":"https://pith.science/pith/YWGZE6AI4C2FDBL7RFIK2M5AU7.json","view_paper":"https://pith.science/paper/YWGZE6AI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2510.00419&json=true","fetch_graph":"https://pith.science/api/pith-number/YWGZE6AI4C2FDBL7RFIK2M5AU7/graph.json","fetch_events":"https://pith.science/api/pith-number/YWGZE6AI4C2FDBL7RFIK2M5AU7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YWGZE6AI4C2FDBL7RFIK2M5AU7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YWGZE6AI4C2FDBL7RFIK2M5AU7/action/storage_attestation","attest_author":"https://pith.science/pith/YWGZE6AI4C2FDBL7RFIK2M5AU7/action/author_attestation","sign_citation":"https://pith.science/pith/YWGZE6AI4C2FDBL7RFIK2M5AU7/action/citation_signature","submit_replication":"https://pith.science/pith/YWGZE6AI4C2FDBL7RFIK2M5AU7/action/replication_record"}},"created_at":"2026-06-01T01:02:22.073023+00:00","updated_at":"2026-06-01T01:02:22.073023+00:00"}