{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:XJB5NQ7CGREZC75SZUYYQL5VGZ","short_pith_number":"pith:XJB5NQ7C","schema_version":"1.0","canonical_sha256":"ba43d6c3e23449917fb2cd31882fb5364fc9c8a2e5a2a75377fc6e141d5ebc8c","source":{"kind":"arxiv","id":"2403.03375","version":3},"attestation_state":"computed","paper":{"title":"Complexity Matters: Dynamics of Feature Learning in the Presence of Spurious Correlations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Da Kuang, Guanwen Qiu, Surbhi Goel","submitted_at":"2024-03-05T23:54:00Z","abstract_excerpt":"Existing research often posits spurious features as easier to learn than core features in neural network optimization, but the impact of their relative simplicity remains under-explored. Moreover, studies mainly focus on end performance rather than the learning dynamics of feature learning. In this paper, we propose a theoretical framework and an associated synthetic dataset grounded in boolean function analysis. This setup allows for fine-grained control over the relative complexity (compared to core features) and correlation strength (with respect to the label) of spurious features to study "},"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":"2403.03375","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-03-05T23:54:00Z","cross_cats_sorted":[],"title_canon_sha256":"74fcb9b3dd576075ef8f7ab459f94c630780946e41f865ec06d1a7b69947a915","abstract_canon_sha256":"d18c5479a752d5f19c228178c33a37afd188800f2d87d0c837fc78ac799e6162"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:58:42.796057Z","signature_b64":"SSAURDM9D2VSchaCzCo7CIL6lhI2uEk7JB1MJ975Aw1iFPXivGkJ9zfbPJ+A17aRWGwpZFnaKwIUM4oTi3A/BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ba43d6c3e23449917fb2cd31882fb5364fc9c8a2e5a2a75377fc6e141d5ebc8c","last_reissued_at":"2026-07-05T08:58:42.795581Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:58:42.795581Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Complexity Matters: Dynamics of Feature Learning in the Presence of Spurious Correlations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Da Kuang, Guanwen Qiu, Surbhi Goel","submitted_at":"2024-03-05T23:54:00Z","abstract_excerpt":"Existing research often posits spurious features as easier to learn than core features in neural network optimization, but the impact of their relative simplicity remains under-explored. Moreover, studies mainly focus on end performance rather than the learning dynamics of feature learning. In this paper, we propose a theoretical framework and an associated synthetic dataset grounded in boolean function analysis. This setup allows for fine-grained control over the relative complexity (compared to core features) and correlation strength (with respect to the label) of spurious features to study "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.03375","kind":"arxiv","version":3},"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/2403.03375/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":"2403.03375","created_at":"2026-07-05T08:58:42.795637+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.03375v3","created_at":"2026-07-05T08:58:42.795637+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.03375","created_at":"2026-07-05T08:58:42.795637+00:00"},{"alias_kind":"pith_short_12","alias_value":"XJB5NQ7CGREZ","created_at":"2026-07-05T08:58:42.795637+00:00"},{"alias_kind":"pith_short_16","alias_value":"XJB5NQ7CGREZC75S","created_at":"2026-07-05T08:58:42.795637+00:00"},{"alias_kind":"pith_short_8","alias_value":"XJB5NQ7C","created_at":"2026-07-05T08:58:42.795637+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.02830","citing_title":"Mitigating Spurious Correlations with Memorization-Guided Dataset De-Biasing","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2411.05183","citing_title":"Why CNN Features Are not Gaussian: A Statistical Anatomy of Deep Representations","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11134","citing_title":"Spurious Correlation Learning in Preference Optimization: Mechanisms, Consequences, and Mitigation via Tie Training","ref_index":16,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XJB5NQ7CGREZC75SZUYYQL5VGZ","json":"https://pith.science/pith/XJB5NQ7CGREZC75SZUYYQL5VGZ.json","graph_json":"https://pith.science/api/pith-number/XJB5NQ7CGREZC75SZUYYQL5VGZ/graph.json","events_json":"https://pith.science/api/pith-number/XJB5NQ7CGREZC75SZUYYQL5VGZ/events.json","paper":"https://pith.science/paper/XJB5NQ7C"},"agent_actions":{"view_html":"https://pith.science/pith/XJB5NQ7CGREZC75SZUYYQL5VGZ","download_json":"https://pith.science/pith/XJB5NQ7CGREZC75SZUYYQL5VGZ.json","view_paper":"https://pith.science/paper/XJB5NQ7C","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.03375&json=true","fetch_graph":"https://pith.science/api/pith-number/XJB5NQ7CGREZC75SZUYYQL5VGZ/graph.json","fetch_events":"https://pith.science/api/pith-number/XJB5NQ7CGREZC75SZUYYQL5VGZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XJB5NQ7CGREZC75SZUYYQL5VGZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XJB5NQ7CGREZC75SZUYYQL5VGZ/action/storage_attestation","attest_author":"https://pith.science/pith/XJB5NQ7CGREZC75SZUYYQL5VGZ/action/author_attestation","sign_citation":"https://pith.science/pith/XJB5NQ7CGREZC75SZUYYQL5VGZ/action/citation_signature","submit_replication":"https://pith.science/pith/XJB5NQ7CGREZC75SZUYYQL5VGZ/action/replication_record"}},"created_at":"2026-07-05T08:58:42.795637+00:00","updated_at":"2026-07-05T08:58:42.795637+00:00"}