{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:P5EYUV2HXTXLVF4UEQFQDPQEYU","short_pith_number":"pith:P5EYUV2H","schema_version":"1.0","canonical_sha256":"7f498a5747bceeba9794240b01be04c5340f1720ff1e406ede235fd38c7e0c02","source":{"kind":"arxiv","id":"2203.14960","version":3},"attestation_state":"computed","paper":{"title":"Domino: Discovering Systematic Errors with Cross-Modal Embeddings","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Christopher Lee-Messer, Christopher R\\'e, James Zou, Jared Dunnmon, Jean-Benoit Delbrouck, Khaled Saab, Maya Varma, Sabri Eyuboglu","submitted_at":"2022-03-24T22:38:56Z","abstract_excerpt":"Machine learning models that achieve high overall accuracy often make systematic errors on important subsets (or slices) of data. Identifying underperforming slices is particularly challenging when working with high-dimensional inputs (e.g. images, audio), where important slices are often unlabeled. In order to address this issue, recent studies have proposed automated slice discovery methods (SDMs), which leverage learned model representations to mine input data for slices on which a model performs poorly. To be useful to a practitioner, these methods must identify slices that are both underp"},"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":"2203.14960","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-03-24T22:38:56Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"6d92ca24020c5717c074bd4020cd9eb764a709f3680d25239c1810f684ddbeed","abstract_canon_sha256":"a0355636881165fca277ed515b6383260e1f923cf9b3abb260b6343e189d05cb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:25:16.715890Z","signature_b64":"P5Y5MRo+jIou+Zarr/+J+/6aHc5uqgA1B1ok+qSVYq/EszhND6k+PDlYWMYJzMDFnYMmCSMXPHN5xXszYs5nAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7f498a5747bceeba9794240b01be04c5340f1720ff1e406ede235fd38c7e0c02","last_reissued_at":"2026-07-05T04:25:16.715442Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:25:16.715442Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Domino: Discovering Systematic Errors with Cross-Modal Embeddings","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Christopher Lee-Messer, Christopher R\\'e, James Zou, Jared Dunnmon, Jean-Benoit Delbrouck, Khaled Saab, Maya Varma, Sabri Eyuboglu","submitted_at":"2022-03-24T22:38:56Z","abstract_excerpt":"Machine learning models that achieve high overall accuracy often make systematic errors on important subsets (or slices) of data. Identifying underperforming slices is particularly challenging when working with high-dimensional inputs (e.g. images, audio), where important slices are often unlabeled. In order to address this issue, recent studies have proposed automated slice discovery methods (SDMs), which leverage learned model representations to mine input data for slices on which a model performs poorly. To be useful to a practitioner, these methods must identify slices that are both underp"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.14960","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/2203.14960/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":"2203.14960","created_at":"2026-07-05T04:25:16.715499+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.14960v3","created_at":"2026-07-05T04:25:16.715499+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.14960","created_at":"2026-07-05T04:25:16.715499+00:00"},{"alias_kind":"pith_short_12","alias_value":"P5EYUV2HXTXL","created_at":"2026-07-05T04:25:16.715499+00:00"},{"alias_kind":"pith_short_16","alias_value":"P5EYUV2HXTXLVF4U","created_at":"2026-07-05T04:25:16.715499+00:00"},{"alias_kind":"pith_short_8","alias_value":"P5EYUV2H","created_at":"2026-07-05T04:25:16.715499+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.01624","citing_title":"What to Test Next: Interpretable Coverage Gap Discovery in Driving VLMs","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2605.29827","citing_title":"Fairness Beyond Demographics: Optimizing Performance Across Appearance-Based Hidden Cohorts in Medical Imaging","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02942","citing_title":"Intersectional Disentangling of Temporal and Acquisition Bias in Fetal Ultrasound","ref_index":4,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/P5EYUV2HXTXLVF4UEQFQDPQEYU","json":"https://pith.science/pith/P5EYUV2HXTXLVF4UEQFQDPQEYU.json","graph_json":"https://pith.science/api/pith-number/P5EYUV2HXTXLVF4UEQFQDPQEYU/graph.json","events_json":"https://pith.science/api/pith-number/P5EYUV2HXTXLVF4UEQFQDPQEYU/events.json","paper":"https://pith.science/paper/P5EYUV2H"},"agent_actions":{"view_html":"https://pith.science/pith/P5EYUV2HXTXLVF4UEQFQDPQEYU","download_json":"https://pith.science/pith/P5EYUV2HXTXLVF4UEQFQDPQEYU.json","view_paper":"https://pith.science/paper/P5EYUV2H","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.14960&json=true","fetch_graph":"https://pith.science/api/pith-number/P5EYUV2HXTXLVF4UEQFQDPQEYU/graph.json","fetch_events":"https://pith.science/api/pith-number/P5EYUV2HXTXLVF4UEQFQDPQEYU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/P5EYUV2HXTXLVF4UEQFQDPQEYU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/P5EYUV2HXTXLVF4UEQFQDPQEYU/action/storage_attestation","attest_author":"https://pith.science/pith/P5EYUV2HXTXLVF4UEQFQDPQEYU/action/author_attestation","sign_citation":"https://pith.science/pith/P5EYUV2HXTXLVF4UEQFQDPQEYU/action/citation_signature","submit_replication":"https://pith.science/pith/P5EYUV2HXTXLVF4UEQFQDPQEYU/action/replication_record"}},"created_at":"2026-07-05T04:25:16.715499+00:00","updated_at":"2026-07-05T04:25:16.715499+00:00"}