{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:6XUO7SVWH4TZGLRIXAKCTQ6HKI","short_pith_number":"pith:6XUO7SVW","schema_version":"1.0","canonical_sha256":"f5e8efcab63f27932e28b81429c3c752050a5d5f3acded036435d682d09927bd","source":{"kind":"arxiv","id":"2204.06863","version":4},"attestation_state":"computed","paper":{"title":"ULF: Unsupervised Labeling Function Correction using Cross-Validation for Weak Supervision","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Anastasiia Sedova, Benjamin Roth","submitted_at":"2022-04-14T10:29:01Z","abstract_excerpt":"A cost-effective alternative to manual data labeling is weak supervision (WS), where data samples are automatically annotated using a predefined set of labeling functions (LFs), rule-based mechanisms that generate artificial labels for the associated classes. In this work, we investigate noise reduction techniques for WS based on the principle of k-fold cross-validation. We introduce a new algorithm ULF for Unsupervised Labeling Function correction, which denoises WS data by leveraging models trained on all but some LFs to identify and correct biases specific to the held-out LFs. Specifically,"},"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":"2204.06863","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-04-14T10:29:01Z","cross_cats_sorted":[],"title_canon_sha256":"80fbf428a0884a2660afd833894a9fee96c7f965594e4614a6e04eb5a7ede33d","abstract_canon_sha256":"ccd1ce005d559b84f1b99f9788e3786b4045900c0307da16b77df77144f6a3f3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:30:00.178722Z","signature_b64":"gunmNeDkkzXDuXgp/QVKhPouFt5MkThAcM0f2rg3Sl/iC9acSbRIrCrEcMeVT2zNtDvvxq038Pv2/ggRFuSCCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f5e8efcab63f27932e28b81429c3c752050a5d5f3acded036435d682d09927bd","last_reissued_at":"2026-07-05T07:30:00.178332Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:30:00.178332Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ULF: Unsupervised Labeling Function Correction using Cross-Validation for Weak Supervision","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Anastasiia Sedova, Benjamin Roth","submitted_at":"2022-04-14T10:29:01Z","abstract_excerpt":"A cost-effective alternative to manual data labeling is weak supervision (WS), where data samples are automatically annotated using a predefined set of labeling functions (LFs), rule-based mechanisms that generate artificial labels for the associated classes. In this work, we investigate noise reduction techniques for WS based on the principle of k-fold cross-validation. We introduce a new algorithm ULF for Unsupervised Labeling Function correction, which denoises WS data by leveraging models trained on all but some LFs to identify and correct biases specific to the held-out LFs. Specifically,"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2204.06863","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/2204.06863/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":"2204.06863","created_at":"2026-07-05T07:30:00.178389+00:00"},{"alias_kind":"arxiv_version","alias_value":"2204.06863v4","created_at":"2026-07-05T07:30:00.178389+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2204.06863","created_at":"2026-07-05T07:30:00.178389+00:00"},{"alias_kind":"pith_short_12","alias_value":"6XUO7SVWH4TZ","created_at":"2026-07-05T07:30:00.178389+00:00"},{"alias_kind":"pith_short_16","alias_value":"6XUO7SVWH4TZGLRI","created_at":"2026-07-05T07:30:00.178389+00:00"},{"alias_kind":"pith_short_8","alias_value":"6XUO7SVW","created_at":"2026-07-05T07:30:00.178389+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.23470","citing_title":"Refining Labeling Functions with Limited Labeled Data","ref_index":28,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6XUO7SVWH4TZGLRIXAKCTQ6HKI","json":"https://pith.science/pith/6XUO7SVWH4TZGLRIXAKCTQ6HKI.json","graph_json":"https://pith.science/api/pith-number/6XUO7SVWH4TZGLRIXAKCTQ6HKI/graph.json","events_json":"https://pith.science/api/pith-number/6XUO7SVWH4TZGLRIXAKCTQ6HKI/events.json","paper":"https://pith.science/paper/6XUO7SVW"},"agent_actions":{"view_html":"https://pith.science/pith/6XUO7SVWH4TZGLRIXAKCTQ6HKI","download_json":"https://pith.science/pith/6XUO7SVWH4TZGLRIXAKCTQ6HKI.json","view_paper":"https://pith.science/paper/6XUO7SVW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2204.06863&json=true","fetch_graph":"https://pith.science/api/pith-number/6XUO7SVWH4TZGLRIXAKCTQ6HKI/graph.json","fetch_events":"https://pith.science/api/pith-number/6XUO7SVWH4TZGLRIXAKCTQ6HKI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6XUO7SVWH4TZGLRIXAKCTQ6HKI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6XUO7SVWH4TZGLRIXAKCTQ6HKI/action/storage_attestation","attest_author":"https://pith.science/pith/6XUO7SVWH4TZGLRIXAKCTQ6HKI/action/author_attestation","sign_citation":"https://pith.science/pith/6XUO7SVWH4TZGLRIXAKCTQ6HKI/action/citation_signature","submit_replication":"https://pith.science/pith/6XUO7SVWH4TZGLRIXAKCTQ6HKI/action/replication_record"}},"created_at":"2026-07-05T07:30:00.178389+00:00","updated_at":"2026-07-05T07:30:00.178389+00:00"}