{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:7EOKS72JTLFBPQBFG6YTE5LWBT","short_pith_number":"pith:7EOKS72J","schema_version":"1.0","canonical_sha256":"f91ca97f499aca17c02537b13275760cf5d45a9be34057f27d5822a3cefcba44","source":{"kind":"arxiv","id":"2405.02596","version":1},"attestation_state":"computed","paper":{"title":"Random Masking Finds Winning Tickets for Parameter Efficient Fine-tuning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Jing Xu, Jingzhao Zhang","submitted_at":"2024-05-04T07:44:18Z","abstract_excerpt":"Fine-tuning large language models (LLM) can be costly. Parameter-efficient fine-tuning (PEFT) addresses the problems by training a fraction of the parameters, whose success reveals the expressiveness and flexibility of pretrained models. This paper studies the limit of PEFT, by further simplifying its design and reducing the number of trainable parameters beyond standard setups. To this end, we use Random Masking to fine-tune the pretrained model. Despite its simplicity, we show that Random Masking is surprisingly effective: with a larger-than-expected learning rate, Random Masking can match t"},"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":"2405.02596","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-05-04T07:44:18Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"680ea485fb42121199a7470979b6368a591149cee351583d458232b922fa9fa8","abstract_canon_sha256":"6ce4332e5d8fd6a1ab354e3c9e84a598deba9553e6ead706adea0d48e7d06ada"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:15:17.867560Z","signature_b64":"az/Njt4tJnHu6QhABLH7iNuAQjaKdCrjsZKQB05I0bP4Zt/BNP9t53dRNzp3IF+ZJsELeCORZxoQe3sP+ArxCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f91ca97f499aca17c02537b13275760cf5d45a9be34057f27d5822a3cefcba44","last_reissued_at":"2026-07-05T08:15:17.867182Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:15:17.867182Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Random Masking Finds Winning Tickets for Parameter Efficient Fine-tuning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Jing Xu, Jingzhao Zhang","submitted_at":"2024-05-04T07:44:18Z","abstract_excerpt":"Fine-tuning large language models (LLM) can be costly. Parameter-efficient fine-tuning (PEFT) addresses the problems by training a fraction of the parameters, whose success reveals the expressiveness and flexibility of pretrained models. This paper studies the limit of PEFT, by further simplifying its design and reducing the number of trainable parameters beyond standard setups. To this end, we use Random Masking to fine-tune the pretrained model. Despite its simplicity, we show that Random Masking is surprisingly effective: with a larger-than-expected learning rate, Random Masking can match t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.02596","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/2405.02596/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":"2405.02596","created_at":"2026-07-05T08:15:17.867249+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.02596v1","created_at":"2026-07-05T08:15:17.867249+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.02596","created_at":"2026-07-05T08:15:17.867249+00:00"},{"alias_kind":"pith_short_12","alias_value":"7EOKS72JTLFB","created_at":"2026-07-05T08:15:17.867249+00:00"},{"alias_kind":"pith_short_16","alias_value":"7EOKS72JTLFBPQBF","created_at":"2026-07-05T08:15:17.867249+00:00"},{"alias_kind":"pith_short_8","alias_value":"7EOKS72J","created_at":"2026-07-05T08:15:17.867249+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.18911","citing_title":"Enhancing Model Privacy in Federated Learning with Random Masking and Quantization","ref_index":43,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7EOKS72JTLFBPQBFG6YTE5LWBT","json":"https://pith.science/pith/7EOKS72JTLFBPQBFG6YTE5LWBT.json","graph_json":"https://pith.science/api/pith-number/7EOKS72JTLFBPQBFG6YTE5LWBT/graph.json","events_json":"https://pith.science/api/pith-number/7EOKS72JTLFBPQBFG6YTE5LWBT/events.json","paper":"https://pith.science/paper/7EOKS72J"},"agent_actions":{"view_html":"https://pith.science/pith/7EOKS72JTLFBPQBFG6YTE5LWBT","download_json":"https://pith.science/pith/7EOKS72JTLFBPQBFG6YTE5LWBT.json","view_paper":"https://pith.science/paper/7EOKS72J","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.02596&json=true","fetch_graph":"https://pith.science/api/pith-number/7EOKS72JTLFBPQBFG6YTE5LWBT/graph.json","fetch_events":"https://pith.science/api/pith-number/7EOKS72JTLFBPQBFG6YTE5LWBT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7EOKS72JTLFBPQBFG6YTE5LWBT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7EOKS72JTLFBPQBFG6YTE5LWBT/action/storage_attestation","attest_author":"https://pith.science/pith/7EOKS72JTLFBPQBFG6YTE5LWBT/action/author_attestation","sign_citation":"https://pith.science/pith/7EOKS72JTLFBPQBFG6YTE5LWBT/action/citation_signature","submit_replication":"https://pith.science/pith/7EOKS72JTLFBPQBFG6YTE5LWBT/action/replication_record"}},"created_at":"2026-07-05T08:15:17.867249+00:00","updated_at":"2026-07-05T08:15:17.867249+00:00"}