{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:KWNJQAOT62HS5NW2XYR6BN6UGA","short_pith_number":"pith:KWNJQAOT","schema_version":"1.0","canonical_sha256":"559a9801d3f68f2eb6dabe23e0b7d43017865a8fb685f0ed5b32ca42c3e38844","source":{"kind":"arxiv","id":"2605.21455","version":1},"attestation_state":"computed","paper":{"title":"Mitigating Label Bias with Interpretable Rubric Embeddings","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Calvin Isley, Johann D. Gaebler, Sharad Goel","submitted_at":"2026-05-20T17:46:08Z","abstract_excerpt":"Statistical decision algorithms are increasingly deployed in domains where ground-truth labels are hard to obtain, such as hiring, university admissions, and content moderation. In these settings, models are typically trained on historical human evaluations -- for example, using past hiring decisions as a proxy for true applicant quality. However, if past evaluations unjustly favor certain groups, models trained on these labels may inherit those biases. To address this problem, we propose basing predictions on rubric embeddings, a representation framework that replaces standard black-box embed"},"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":"2605.21455","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-20T17:46:08Z","cross_cats_sorted":[],"title_canon_sha256":"440d8a9afa982c58ac4174b272fcf4677a9a22d24d934fcdf98252e77a3c2dcf","abstract_canon_sha256":"ce7aa3e13908e0ec05354ca6c52330ff0ef6530dce0a62d4b70136e930c0dd64"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-21T02:05:38.070166Z","signature_b64":"R8+RG8UtmihvmcLikqyIVWc2b/DYSHfKlJyoYBaYjCBqh14Ng4BWe+IlDaNfBsVyJHzf9x2qMNHwlagFB6x/Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"559a9801d3f68f2eb6dabe23e0b7d43017865a8fb685f0ed5b32ca42c3e38844","last_reissued_at":"2026-05-21T02:05:38.069422Z","signature_status":"signed_v1","first_computed_at":"2026-05-21T02:05:38.069422Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Mitigating Label Bias with Interpretable Rubric Embeddings","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Calvin Isley, Johann D. Gaebler, Sharad Goel","submitted_at":"2026-05-20T17:46:08Z","abstract_excerpt":"Statistical decision algorithms are increasingly deployed in domains where ground-truth labels are hard to obtain, such as hiring, university admissions, and content moderation. In these settings, models are typically trained on historical human evaluations -- for example, using past hiring decisions as a proxy for true applicant quality. However, if past evaluations unjustly favor certain groups, models trained on these labels may inherit those biases. To address this problem, we propose basing predictions on rubric embeddings, a representation framework that replaces standard black-box embed"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.21455","kind":"arxiv","version":1},"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/2605.21455/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":"2605.21455","created_at":"2026-05-21T02:05:38.069538+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.21455v1","created_at":"2026-05-21T02:05:38.069538+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.21455","created_at":"2026-05-21T02:05:38.069538+00:00"},{"alias_kind":"pith_short_12","alias_value":"KWNJQAOT62HS","created_at":"2026-05-21T02:05:38.069538+00:00"},{"alias_kind":"pith_short_16","alias_value":"KWNJQAOT62HS5NW2","created_at":"2026-05-21T02:05:38.069538+00:00"},{"alias_kind":"pith_short_8","alias_value":"KWNJQAOT","created_at":"2026-05-21T02:05:38.069538+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/KWNJQAOT62HS5NW2XYR6BN6UGA","json":"https://pith.science/pith/KWNJQAOT62HS5NW2XYR6BN6UGA.json","graph_json":"https://pith.science/api/pith-number/KWNJQAOT62HS5NW2XYR6BN6UGA/graph.json","events_json":"https://pith.science/api/pith-number/KWNJQAOT62HS5NW2XYR6BN6UGA/events.json","paper":"https://pith.science/paper/KWNJQAOT"},"agent_actions":{"view_html":"https://pith.science/pith/KWNJQAOT62HS5NW2XYR6BN6UGA","download_json":"https://pith.science/pith/KWNJQAOT62HS5NW2XYR6BN6UGA.json","view_paper":"https://pith.science/paper/KWNJQAOT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.21455&json=true","fetch_graph":"https://pith.science/api/pith-number/KWNJQAOT62HS5NW2XYR6BN6UGA/graph.json","fetch_events":"https://pith.science/api/pith-number/KWNJQAOT62HS5NW2XYR6BN6UGA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KWNJQAOT62HS5NW2XYR6BN6UGA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KWNJQAOT62HS5NW2XYR6BN6UGA/action/storage_attestation","attest_author":"https://pith.science/pith/KWNJQAOT62HS5NW2XYR6BN6UGA/action/author_attestation","sign_citation":"https://pith.science/pith/KWNJQAOT62HS5NW2XYR6BN6UGA/action/citation_signature","submit_replication":"https://pith.science/pith/KWNJQAOT62HS5NW2XYR6BN6UGA/action/replication_record"}},"created_at":"2026-05-21T02:05:38.069538+00:00","updated_at":"2026-05-21T02:05:38.069538+00:00"}