{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:2R2XAXW77XPETUMSMTR4J6E2CD","short_pith_number":"pith:2R2XAXW7","schema_version":"1.0","canonical_sha256":"d475705edffdde49d19264e3c4f89a10f285619d203bae3c3e48ca0a55cc2fa1","source":{"kind":"arxiv","id":"2105.06422","version":3},"attestation_state":"computed","paper":{"title":"Causally motivated Shortcut Removal Using Auxiliary Labels","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alexander D'Amour, Ben Packer, Dan Moldovan, Davis Blalock, Maggie Makar, Yoni Halpern","submitted_at":"2021-05-13T16:58:45Z","abstract_excerpt":"Shortcut learning, in which models make use of easy-to-represent but unstable associations, is a major failure mode for robust machine learning. We study a flexible, causally-motivated approach to training robust predictors by discouraging the use of specific shortcuts, focusing on a common setting where a robust predictor could achieve optimal \\emph{iid} generalization in principle, but is overshadowed by a shortcut predictor in practice. Our approach uses auxiliary labels, typically available at training time, to enforce conditional independences implied by the causal graph. We show both the"},"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":"2105.06422","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-05-13T16:58:45Z","cross_cats_sorted":[],"title_canon_sha256":"223357215d6cc164a59e55e4439b9285fba279e091a465ff156de7159c527f99","abstract_canon_sha256":"64729140783dbe7cbf3e747d224c46a8b7b4b06dc9765f57cd56e73d1718e33c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:59:22.075827Z","signature_b64":"q1d5eWCi9tTbuxQ7ibDU10uqyk4sZh3uLRg/K4R/GHlf9+S+RD5hLDNtnyFY4eTMLGb2muqdcd64BX0LKTuPBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d475705edffdde49d19264e3c4f89a10f285619d203bae3c3e48ca0a55cc2fa1","last_reissued_at":"2026-07-05T03:59:22.075350Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:59:22.075350Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Causally motivated Shortcut Removal Using Auxiliary Labels","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alexander D'Amour, Ben Packer, Dan Moldovan, Davis Blalock, Maggie Makar, Yoni Halpern","submitted_at":"2021-05-13T16:58:45Z","abstract_excerpt":"Shortcut learning, in which models make use of easy-to-represent but unstable associations, is a major failure mode for robust machine learning. We study a flexible, causally-motivated approach to training robust predictors by discouraging the use of specific shortcuts, focusing on a common setting where a robust predictor could achieve optimal \\emph{iid} generalization in principle, but is overshadowed by a shortcut predictor in practice. Our approach uses auxiliary labels, typically available at training time, to enforce conditional independences implied by the causal graph. We show both the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2105.06422","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/2105.06422/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":"2105.06422","created_at":"2026-07-05T03:59:22.075408+00:00"},{"alias_kind":"arxiv_version","alias_value":"2105.06422v3","created_at":"2026-07-05T03:59:22.075408+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2105.06422","created_at":"2026-07-05T03:59:22.075408+00:00"},{"alias_kind":"pith_short_12","alias_value":"2R2XAXW77XPE","created_at":"2026-07-05T03:59:22.075408+00:00"},{"alias_kind":"pith_short_16","alias_value":"2R2XAXW77XPETUMS","created_at":"2026-07-05T03:59:22.075408+00:00"},{"alias_kind":"pith_short_8","alias_value":"2R2XAXW7","created_at":"2026-07-05T03:59:22.075408+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.18878","citing_title":"Prognostic Value of Lung Ultrasound Biomarkers for Readmission Risk in Congestive Heart Failure: A Pilot Data-Driven Analysis","ref_index":195,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2R2XAXW77XPETUMSMTR4J6E2CD","json":"https://pith.science/pith/2R2XAXW77XPETUMSMTR4J6E2CD.json","graph_json":"https://pith.science/api/pith-number/2R2XAXW77XPETUMSMTR4J6E2CD/graph.json","events_json":"https://pith.science/api/pith-number/2R2XAXW77XPETUMSMTR4J6E2CD/events.json","paper":"https://pith.science/paper/2R2XAXW7"},"agent_actions":{"view_html":"https://pith.science/pith/2R2XAXW77XPETUMSMTR4J6E2CD","download_json":"https://pith.science/pith/2R2XAXW77XPETUMSMTR4J6E2CD.json","view_paper":"https://pith.science/paper/2R2XAXW7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2105.06422&json=true","fetch_graph":"https://pith.science/api/pith-number/2R2XAXW77XPETUMSMTR4J6E2CD/graph.json","fetch_events":"https://pith.science/api/pith-number/2R2XAXW77XPETUMSMTR4J6E2CD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2R2XAXW77XPETUMSMTR4J6E2CD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2R2XAXW77XPETUMSMTR4J6E2CD/action/storage_attestation","attest_author":"https://pith.science/pith/2R2XAXW77XPETUMSMTR4J6E2CD/action/author_attestation","sign_citation":"https://pith.science/pith/2R2XAXW77XPETUMSMTR4J6E2CD/action/citation_signature","submit_replication":"https://pith.science/pith/2R2XAXW77XPETUMSMTR4J6E2CD/action/replication_record"}},"created_at":"2026-07-05T03:59:22.075408+00:00","updated_at":"2026-07-05T03:59:22.075408+00:00"}