{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:JLB7G6CIY44P34TZEDCBXIHUJY","short_pith_number":"pith:JLB7G6CI","schema_version":"1.0","canonical_sha256":"4ac3f37848c738fdf27920c41ba0f44e20719fbf66ee5d8713f2bc5d73b3ad49","source":{"kind":"arxiv","id":"2602.00827","version":2},"attestation_state":"computed","paper":{"title":"Over-Alignment vs Over-Fitting: The Role of Feature Learning Strength in Generalization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Jaeho Lee, Taehyeok Ha, Taesun Yeom","submitted_at":"2026-01-31T17:43:02Z","abstract_excerpt":"Feature learning strength (FLS), i.e., the inverse of the effective output scaling of a model, plays a critical role in shaping the optimization dynamics of neural nets. While its impact has been extensively studied under the asymptotic regimes -- both in training time and FLS -- existing theory offers limited insight into how FLS affects generalization in practical settings, such as when training is stopped upon reaching a target training risk. In this work, we investigate the impact of FLS on generalization in deep networks under such practical conditions. Through empirical studies, we first"},"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":"2602.00827","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-01-31T17:43:02Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"9ccec6f81b090bd7c1a6ccd69f9bf1363ca05dd77e0645a5e8dcfd51921627a3","abstract_canon_sha256":"4ad905c285d61cde2cba1b214916679a1988e0bf11771deaeaa9b899db40610b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-27T01:04:55.660557Z","signature_b64":"YJwqPYUEXFwCT/IozP94tdMnCVC5wgerJkm+pKXUkOT7t+B1ULiaDr+4vlfFImrmdngewJND2dkQnULlAWsWDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4ac3f37848c738fdf27920c41ba0f44e20719fbf66ee5d8713f2bc5d73b3ad49","last_reissued_at":"2026-05-27T01:04:55.659668Z","signature_status":"signed_v1","first_computed_at":"2026-05-27T01:04:55.659668Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Over-Alignment vs Over-Fitting: The Role of Feature Learning Strength in Generalization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Jaeho Lee, Taehyeok Ha, Taesun Yeom","submitted_at":"2026-01-31T17:43:02Z","abstract_excerpt":"Feature learning strength (FLS), i.e., the inverse of the effective output scaling of a model, plays a critical role in shaping the optimization dynamics of neural nets. While its impact has been extensively studied under the asymptotic regimes -- both in training time and FLS -- existing theory offers limited insight into how FLS affects generalization in practical settings, such as when training is stopped upon reaching a target training risk. In this work, we investigate the impact of FLS on generalization in deep networks under such practical conditions. Through empirical studies, we first"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.00827","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/2602.00827/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":"2602.00827","created_at":"2026-05-27T01:04:55.659778+00:00"},{"alias_kind":"arxiv_version","alias_value":"2602.00827v2","created_at":"2026-05-27T01:04:55.659778+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.00827","created_at":"2026-05-27T01:04:55.659778+00:00"},{"alias_kind":"pith_short_12","alias_value":"JLB7G6CIY44P","created_at":"2026-05-27T01:04:55.659778+00:00"},{"alias_kind":"pith_short_16","alias_value":"JLB7G6CIY44P34TZ","created_at":"2026-05-27T01:04:55.659778+00:00"},{"alias_kind":"pith_short_8","alias_value":"JLB7G6CI","created_at":"2026-05-27T01:04:55.659778+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/JLB7G6CIY44P34TZEDCBXIHUJY","json":"https://pith.science/pith/JLB7G6CIY44P34TZEDCBXIHUJY.json","graph_json":"https://pith.science/api/pith-number/JLB7G6CIY44P34TZEDCBXIHUJY/graph.json","events_json":"https://pith.science/api/pith-number/JLB7G6CIY44P34TZEDCBXIHUJY/events.json","paper":"https://pith.science/paper/JLB7G6CI"},"agent_actions":{"view_html":"https://pith.science/pith/JLB7G6CIY44P34TZEDCBXIHUJY","download_json":"https://pith.science/pith/JLB7G6CIY44P34TZEDCBXIHUJY.json","view_paper":"https://pith.science/paper/JLB7G6CI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2602.00827&json=true","fetch_graph":"https://pith.science/api/pith-number/JLB7G6CIY44P34TZEDCBXIHUJY/graph.json","fetch_events":"https://pith.science/api/pith-number/JLB7G6CIY44P34TZEDCBXIHUJY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JLB7G6CIY44P34TZEDCBXIHUJY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JLB7G6CIY44P34TZEDCBXIHUJY/action/storage_attestation","attest_author":"https://pith.science/pith/JLB7G6CIY44P34TZEDCBXIHUJY/action/author_attestation","sign_citation":"https://pith.science/pith/JLB7G6CIY44P34TZEDCBXIHUJY/action/citation_signature","submit_replication":"https://pith.science/pith/JLB7G6CIY44P34TZEDCBXIHUJY/action/replication_record"}},"created_at":"2026-05-27T01:04:55.659778+00:00","updated_at":"2026-05-27T01:04:55.659778+00:00"}