{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:KOQSXFCYJL72WSMXBEFUDGFHJE","short_pith_number":"pith:KOQSXFCY","schema_version":"1.0","canonical_sha256":"53a12b94584affab4997090b4198a7491ab4e36442ab72f82fceef88358a9e3f","source":{"kind":"arxiv","id":"2405.07527","version":1},"attestation_state":"computed","paper":{"title":"Train Faster, Perform Better: Modular Adaptive Training in Over-Parameterized Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Dongsheng Li, Fan Yang, Li Shang, Mingzhi Dong, Ning Gu, Qin Lv, Robert P. Dick, Tun Lu, Xiaochen Yang, Yingying Zhao, Yixuan Chen, Yubin Shi, Yujiang Wang","submitted_at":"2024-05-13T07:46:48Z","abstract_excerpt":"Despite their prevalence in deep-learning communities, over-parameterized models convey high demands of computational costs for proper training. This work studies the fine-grained, modular-level learning dynamics of over-parameterized models to attain a more efficient and fruitful training strategy. Empirical evidence reveals that when scaling down into network modules, such as heads in self-attention models, we can observe varying learning patterns implicitly associated with each module's trainability. To describe such modular-level learning capabilities, we introduce a novel concept dubbed m"},"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":"2405.07527","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-05-13T07:46:48Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ca49ec3ad945e676ca5d82f45f5f3e4a5e57f9e4ccf0c84754d941e38f6be7d4","abstract_canon_sha256":"743404203e769eb981e3e16e869e9d97457fb666a4190ee2bd6ed67d3a97ae89"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:18:24.296332Z","signature_b64":"/8Tf48qiNEIWzfZ9YMib5ig4JhMA5AL4Sll0NKszXtwBsY/6wYfXiT6C9c55DApBzx7LO25+6xwBC6RMvzW6Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"53a12b94584affab4997090b4198a7491ab4e36442ab72f82fceef88358a9e3f","last_reissued_at":"2026-07-05T08:18:24.295860Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:18:24.295860Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Train Faster, Perform Better: Modular Adaptive Training in Over-Parameterized Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Dongsheng Li, Fan Yang, Li Shang, Mingzhi Dong, Ning Gu, Qin Lv, Robert P. Dick, Tun Lu, Xiaochen Yang, Yingying Zhao, Yixuan Chen, Yubin Shi, Yujiang Wang","submitted_at":"2024-05-13T07:46:48Z","abstract_excerpt":"Despite their prevalence in deep-learning communities, over-parameterized models convey high demands of computational costs for proper training. This work studies the fine-grained, modular-level learning dynamics of over-parameterized models to attain a more efficient and fruitful training strategy. Empirical evidence reveals that when scaling down into network modules, such as heads in self-attention models, we can observe varying learning patterns implicitly associated with each module's trainability. To describe such modular-level learning capabilities, we introduce a novel concept dubbed m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.07527","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/2405.07527/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":"2405.07527","created_at":"2026-07-05T08:18:24.295914+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.07527v1","created_at":"2026-07-05T08:18:24.295914+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.07527","created_at":"2026-07-05T08:18:24.295914+00:00"},{"alias_kind":"pith_short_12","alias_value":"KOQSXFCYJL72","created_at":"2026-07-05T08:18:24.295914+00:00"},{"alias_kind":"pith_short_16","alias_value":"KOQSXFCYJL72WSMX","created_at":"2026-07-05T08:18:24.295914+00:00"},{"alias_kind":"pith_short_8","alias_value":"KOQSXFCY","created_at":"2026-07-05T08:18:24.295914+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/KOQSXFCYJL72WSMXBEFUDGFHJE","json":"https://pith.science/pith/KOQSXFCYJL72WSMXBEFUDGFHJE.json","graph_json":"https://pith.science/api/pith-number/KOQSXFCYJL72WSMXBEFUDGFHJE/graph.json","events_json":"https://pith.science/api/pith-number/KOQSXFCYJL72WSMXBEFUDGFHJE/events.json","paper":"https://pith.science/paper/KOQSXFCY"},"agent_actions":{"view_html":"https://pith.science/pith/KOQSXFCYJL72WSMXBEFUDGFHJE","download_json":"https://pith.science/pith/KOQSXFCYJL72WSMXBEFUDGFHJE.json","view_paper":"https://pith.science/paper/KOQSXFCY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.07527&json=true","fetch_graph":"https://pith.science/api/pith-number/KOQSXFCYJL72WSMXBEFUDGFHJE/graph.json","fetch_events":"https://pith.science/api/pith-number/KOQSXFCYJL72WSMXBEFUDGFHJE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KOQSXFCYJL72WSMXBEFUDGFHJE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KOQSXFCYJL72WSMXBEFUDGFHJE/action/storage_attestation","attest_author":"https://pith.science/pith/KOQSXFCYJL72WSMXBEFUDGFHJE/action/author_attestation","sign_citation":"https://pith.science/pith/KOQSXFCYJL72WSMXBEFUDGFHJE/action/citation_signature","submit_replication":"https://pith.science/pith/KOQSXFCYJL72WSMXBEFUDGFHJE/action/replication_record"}},"created_at":"2026-07-05T08:18:24.295914+00:00","updated_at":"2026-07-05T08:18:24.295914+00:00"}