{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:SKHBFLCYY35TCKZDIJ7Y62QXX6","short_pith_number":"pith:SKHBFLCY","schema_version":"1.0","canonical_sha256":"928e12ac58c6fb312b23427f8f6a17bf962170245da8d083268344a75e8b44ef","source":{"kind":"arxiv","id":"2410.14086","version":4},"attestation_state":"computed","paper":{"title":"In-context learning and Occam's razor","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Dhanya Sridhar, Eric Elmoznino, Guillaume Lajoie, Leo Gagnon, Mahan Fathi, Sarthak Mittal, Tejas Kasetty, Tom Marty","submitted_at":"2024-10-17T23:37:34Z","abstract_excerpt":"A central goal of machine learning is generalization. While the No Free Lunch Theorem states that we cannot obtain theoretical guarantees for generalization without further assumptions, in practice we observe that simple models which explain the training data generalize best: a principle called Occam's razor. Despite the need for simple models, most current approaches in machine learning only minimize the training error, and at best indirectly promote simplicity through regularization or architecture design. Here, we draw a connection between Occam's razor and in-context learning: an emergent "},"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":"2410.14086","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-17T23:37:34Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"6e31156eaab217d779407d174283b8fb4124fd1d05aba0e75c207a7028caebb8","abstract_canon_sha256":"31d0b8c0acf694af56c6564247efc139d28c86dd5a97384d196cde739cb01e4f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:14:25.297123Z","signature_b64":"j960T5RX9XhfcMo58YiasavAR3Fg/cYQzFLMC0b9sT7tRPs4AniFdjZ1nDGiSKt/L5Znp16cmAfWhu4z8on1Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"928e12ac58c6fb312b23427f8f6a17bf962170245da8d083268344a75e8b44ef","last_reissued_at":"2026-07-05T11:14:25.296664Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:14:25.296664Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"In-context learning and Occam's razor","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Dhanya Sridhar, Eric Elmoznino, Guillaume Lajoie, Leo Gagnon, Mahan Fathi, Sarthak Mittal, Tejas Kasetty, Tom Marty","submitted_at":"2024-10-17T23:37:34Z","abstract_excerpt":"A central goal of machine learning is generalization. While the No Free Lunch Theorem states that we cannot obtain theoretical guarantees for generalization without further assumptions, in practice we observe that simple models which explain the training data generalize best: a principle called Occam's razor. Despite the need for simple models, most current approaches in machine learning only minimize the training error, and at best indirectly promote simplicity through regularization or architecture design. Here, we draw a connection between Occam's razor and in-context learning: an emergent "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.14086","kind":"arxiv","version":4},"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/2410.14086/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":"2410.14086","created_at":"2026-07-05T11:14:25.296721+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.14086v4","created_at":"2026-07-05T11:14:25.296721+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.14086","created_at":"2026-07-05T11:14:25.296721+00:00"},{"alias_kind":"pith_short_12","alias_value":"SKHBFLCYY35T","created_at":"2026-07-05T11:14:25.296721+00:00"},{"alias_kind":"pith_short_16","alias_value":"SKHBFLCYY35TCKZD","created_at":"2026-07-05T11:14:25.296721+00:00"},{"alias_kind":"pith_short_8","alias_value":"SKHBFLCY","created_at":"2026-07-05T11:14:25.296721+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.09240","citing_title":"Task Vectors in In-Context Learning: Emergence, Formation, and Benefit","ref_index":2021,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SKHBFLCYY35TCKZDIJ7Y62QXX6","json":"https://pith.science/pith/SKHBFLCYY35TCKZDIJ7Y62QXX6.json","graph_json":"https://pith.science/api/pith-number/SKHBFLCYY35TCKZDIJ7Y62QXX6/graph.json","events_json":"https://pith.science/api/pith-number/SKHBFLCYY35TCKZDIJ7Y62QXX6/events.json","paper":"https://pith.science/paper/SKHBFLCY"},"agent_actions":{"view_html":"https://pith.science/pith/SKHBFLCYY35TCKZDIJ7Y62QXX6","download_json":"https://pith.science/pith/SKHBFLCYY35TCKZDIJ7Y62QXX6.json","view_paper":"https://pith.science/paper/SKHBFLCY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.14086&json=true","fetch_graph":"https://pith.science/api/pith-number/SKHBFLCYY35TCKZDIJ7Y62QXX6/graph.json","fetch_events":"https://pith.science/api/pith-number/SKHBFLCYY35TCKZDIJ7Y62QXX6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SKHBFLCYY35TCKZDIJ7Y62QXX6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SKHBFLCYY35TCKZDIJ7Y62QXX6/action/storage_attestation","attest_author":"https://pith.science/pith/SKHBFLCYY35TCKZDIJ7Y62QXX6/action/author_attestation","sign_citation":"https://pith.science/pith/SKHBFLCYY35TCKZDIJ7Y62QXX6/action/citation_signature","submit_replication":"https://pith.science/pith/SKHBFLCYY35TCKZDIJ7Y62QXX6/action/replication_record"}},"created_at":"2026-07-05T11:14:25.296721+00:00","updated_at":"2026-07-05T11:14:25.296721+00:00"}