{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:NZL65IQPORVH43GW54DQT5DFWR","short_pith_number":"pith:NZL65IQP","schema_version":"1.0","canonical_sha256":"6e57eea20f746a7e6cd6ef0709f465b4546cb8383ca181f561a7edee0da267eb","source":{"kind":"arxiv","id":"2410.08194","version":2},"attestation_state":"computed","paper":{"title":"Features are fate: a theory of transfer learning in high-dimensional regression","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Grant M. Rotskoff, Javan Tahir, Surya Ganguli","submitted_at":"2024-10-10T17:58:26Z","abstract_excerpt":"With the emergence of large-scale pre-trained neural networks, methods to adapt such \"foundation\" models to data-limited downstream tasks have become a necessity. Fine-tuning, preference optimization, and transfer learning have all been successfully employed for these purposes when the target task closely resembles the source task, but a precise theoretical understanding of \"task similarity\" is still lacking. While conventional wisdom suggests that simple measures of similarity between source and target distributions, such as $\\phi$-divergences or integral probability metrics, can directly pre"},"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":"2410.08194","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2024-10-10T17:58:26Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"705a2dae9f8382de26caa6a0588c48e7700e5c893f0e0fd1d4669017a103ba93","abstract_canon_sha256":"04f271049297d38e2d7206246097627962d408a7fe4f39e806e983b1a4c2af50"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:33:16.586122Z","signature_b64":"qUpNSbXS6FseDIAxd9HYsS4sfLF8HeNJ6R1bSfu+m/glGO/0KvRB3YnuRrp0rsSQ1kmOFE1ysFh4jZJPF/7LDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6e57eea20f746a7e6cd6ef0709f465b4546cb8383ca181f561a7edee0da267eb","last_reissued_at":"2026-07-05T11:33:16.585601Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:33:16.585601Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Features are fate: a theory of transfer learning in high-dimensional regression","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Grant M. Rotskoff, Javan Tahir, Surya Ganguli","submitted_at":"2024-10-10T17:58:26Z","abstract_excerpt":"With the emergence of large-scale pre-trained neural networks, methods to adapt such \"foundation\" models to data-limited downstream tasks have become a necessity. Fine-tuning, preference optimization, and transfer learning have all been successfully employed for these purposes when the target task closely resembles the source task, but a precise theoretical understanding of \"task similarity\" is still lacking. While conventional wisdom suggests that simple measures of similarity between source and target distributions, such as $\\phi$-divergences or integral probability metrics, can directly pre"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.08194","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/2410.08194/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":"2410.08194","created_at":"2026-07-05T11:33:16.585667+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.08194v2","created_at":"2026-07-05T11:33:16.585667+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.08194","created_at":"2026-07-05T11:33:16.585667+00:00"},{"alias_kind":"pith_short_12","alias_value":"NZL65IQPORVH","created_at":"2026-07-05T11:33:16.585667+00:00"},{"alias_kind":"pith_short_16","alias_value":"NZL65IQPORVH43GW","created_at":"2026-07-05T11:33:16.585667+00:00"},{"alias_kind":"pith_short_8","alias_value":"NZL65IQP","created_at":"2026-07-05T11:33:16.585667+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.23346","citing_title":"Field-level weak lensing cosmology with $<100$ simulations using multifidelity simulation-based inference","ref_index":120,"is_internal_anchor":false},{"citing_arxiv_id":"2606.18333","citing_title":"Polarisation and Faraday rotation measure imaging at metre wavelengths with sub-arcsecond resolution: a foundational calibration strategy","ref_index":9,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NZL65IQPORVH43GW54DQT5DFWR","json":"https://pith.science/pith/NZL65IQPORVH43GW54DQT5DFWR.json","graph_json":"https://pith.science/api/pith-number/NZL65IQPORVH43GW54DQT5DFWR/graph.json","events_json":"https://pith.science/api/pith-number/NZL65IQPORVH43GW54DQT5DFWR/events.json","paper":"https://pith.science/paper/NZL65IQP"},"agent_actions":{"view_html":"https://pith.science/pith/NZL65IQPORVH43GW54DQT5DFWR","download_json":"https://pith.science/pith/NZL65IQPORVH43GW54DQT5DFWR.json","view_paper":"https://pith.science/paper/NZL65IQP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.08194&json=true","fetch_graph":"https://pith.science/api/pith-number/NZL65IQPORVH43GW54DQT5DFWR/graph.json","fetch_events":"https://pith.science/api/pith-number/NZL65IQPORVH43GW54DQT5DFWR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NZL65IQPORVH43GW54DQT5DFWR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NZL65IQPORVH43GW54DQT5DFWR/action/storage_attestation","attest_author":"https://pith.science/pith/NZL65IQPORVH43GW54DQT5DFWR/action/author_attestation","sign_citation":"https://pith.science/pith/NZL65IQPORVH43GW54DQT5DFWR/action/citation_signature","submit_replication":"https://pith.science/pith/NZL65IQPORVH43GW54DQT5DFWR/action/replication_record"}},"created_at":"2026-07-05T11:33:16.585667+00:00","updated_at":"2026-07-05T11:33:16.585667+00:00"}