{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:TXM2RHR5ZPBMLK5P4OIPOLF4VJ","short_pith_number":"pith:TXM2RHR5","schema_version":"1.0","canonical_sha256":"9dd9a89e3dcbc2c5abafe390f72cbcaa482bae16aa659d0daf2477230cce017f","source":{"kind":"arxiv","id":"2502.11513","version":1},"attestation_state":"computed","paper":{"title":"MaZO: Masked Zeroth-Order Optimization for Multi-Task Fine-Tuning of Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Athanasios Mouchtaris, Kai Zhen, Nathan Susanj, Siegfried Kunzmann, Yifan Yang, Zheng Zhang, Zhen Zhang","submitted_at":"2025-02-17T07:28:52Z","abstract_excerpt":"Large language models have demonstrated exceptional capabilities across diverse tasks, but their fine-tuning demands significant memory, posing challenges for resource-constrained environments. Zeroth-order (ZO) optimization provides a memory-efficient alternative by eliminating the need for backpropagation. However, ZO optimization suffers from high gradient variance, and prior research has largely focused on single-task learning, leaving its application to multi-task learning unexplored. Multi-task learning is crucial for leveraging shared knowledge across tasks to improve generalization, ye"},"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":"2502.11513","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-17T07:28:52Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"5eac87250738b98030cd764e2d06c0fdd1fb33d25120ef1dad7cfff7f0cb5bd2","abstract_canon_sha256":"6cdc122855c264e0eeac8d8e42fb3b2536f538bbc610d2809948c67067f59a16"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:15:29.981962Z","signature_b64":"5z/QhwEKEFpdF9Xu8+/NSadXdq82Fk+wGSB0vHq3OZVjcKve+tURIVlwoKLX+w0e8w/Cgxk/HlevFuEdLTgkBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9dd9a89e3dcbc2c5abafe390f72cbcaa482bae16aa659d0daf2477230cce017f","last_reissued_at":"2026-07-05T10:15:29.981470Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:15:29.981470Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MaZO: Masked Zeroth-Order Optimization for Multi-Task Fine-Tuning of Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Athanasios Mouchtaris, Kai Zhen, Nathan Susanj, Siegfried Kunzmann, Yifan Yang, Zheng Zhang, Zhen Zhang","submitted_at":"2025-02-17T07:28:52Z","abstract_excerpt":"Large language models have demonstrated exceptional capabilities across diverse tasks, but their fine-tuning demands significant memory, posing challenges for resource-constrained environments. Zeroth-order (ZO) optimization provides a memory-efficient alternative by eliminating the need for backpropagation. However, ZO optimization suffers from high gradient variance, and prior research has largely focused on single-task learning, leaving its application to multi-task learning unexplored. Multi-task learning is crucial for leveraging shared knowledge across tasks to improve generalization, ye"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.11513","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/2502.11513/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":"2502.11513","created_at":"2026-07-05T10:15:29.981531+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.11513v1","created_at":"2026-07-05T10:15:29.981531+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.11513","created_at":"2026-07-05T10:15:29.981531+00:00"},{"alias_kind":"pith_short_12","alias_value":"TXM2RHR5ZPBM","created_at":"2026-07-05T10:15:29.981531+00:00"},{"alias_kind":"pith_short_16","alias_value":"TXM2RHR5ZPBMLK5P","created_at":"2026-07-05T10:15:29.981531+00:00"},{"alias_kind":"pith_short_8","alias_value":"TXM2RHR5","created_at":"2026-07-05T10:15:29.981531+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/TXM2RHR5ZPBMLK5P4OIPOLF4VJ","json":"https://pith.science/pith/TXM2RHR5ZPBMLK5P4OIPOLF4VJ.json","graph_json":"https://pith.science/api/pith-number/TXM2RHR5ZPBMLK5P4OIPOLF4VJ/graph.json","events_json":"https://pith.science/api/pith-number/TXM2RHR5ZPBMLK5P4OIPOLF4VJ/events.json","paper":"https://pith.science/paper/TXM2RHR5"},"agent_actions":{"view_html":"https://pith.science/pith/TXM2RHR5ZPBMLK5P4OIPOLF4VJ","download_json":"https://pith.science/pith/TXM2RHR5ZPBMLK5P4OIPOLF4VJ.json","view_paper":"https://pith.science/paper/TXM2RHR5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.11513&json=true","fetch_graph":"https://pith.science/api/pith-number/TXM2RHR5ZPBMLK5P4OIPOLF4VJ/graph.json","fetch_events":"https://pith.science/api/pith-number/TXM2RHR5ZPBMLK5P4OIPOLF4VJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TXM2RHR5ZPBMLK5P4OIPOLF4VJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TXM2RHR5ZPBMLK5P4OIPOLF4VJ/action/storage_attestation","attest_author":"https://pith.science/pith/TXM2RHR5ZPBMLK5P4OIPOLF4VJ/action/author_attestation","sign_citation":"https://pith.science/pith/TXM2RHR5ZPBMLK5P4OIPOLF4VJ/action/citation_signature","submit_replication":"https://pith.science/pith/TXM2RHR5ZPBMLK5P4OIPOLF4VJ/action/replication_record"}},"created_at":"2026-07-05T10:15:29.981531+00:00","updated_at":"2026-07-05T10:15:29.981531+00:00"}