{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:R4625KAK2RSPBL4DD4BCGMDQ3D","short_pith_number":"pith:R4625KAK","schema_version":"1.0","canonical_sha256":"8f3daea80ad464f0af831f02233070d8c5f40141028eca41f6158808f00d1e11","source":{"kind":"arxiv","id":"2011.05429","version":1},"attestation_state":"computed","paper":{"title":"Debugging Tests for Model Explanations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Been Kim, Ilaria Liccardi, Julius Adebayo, Michael Muelly","submitted_at":"2020-11-10T22:23:25Z","abstract_excerpt":"We investigate whether post-hoc model explanations are effective for diagnosing model errors--model debugging. In response to the challenge of explaining a model's prediction, a vast array of explanation methods have been proposed. Despite increasing use, it is unclear if they are effective. To start, we categorize \\textit{bugs}, based on their source, into:~\\textit{data, model, and test-time} contamination bugs. For several explanation methods, we assess their ability to: detect spurious correlation artifacts (data contamination), diagnose mislabeled training examples (data contamination), di"},"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":"2011.05429","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2020-11-10T22:23:25Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"191255730334b95f5cddd33f7ce600fe6ae1405b911f004f61d4008bbbe53642","abstract_canon_sha256":"6fb4f2b67d6777e9ee973dfbfbad7e7ce23e8c0cac6bd3ae38aa4eb260471dc2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:50:49.880273Z","signature_b64":"NKL74lKnOw2g8VnpCHx2ZMssDZYfDjlvabtOmq2ZMCsIs3GgkfEko0JzY7s9IZluYbZ/1i7IlMeTqSYJAmORCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8f3daea80ad464f0af831f02233070d8c5f40141028eca41f6158808f00d1e11","last_reissued_at":"2026-07-05T01:50:49.879866Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:50:49.879866Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Debugging Tests for Model Explanations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Been Kim, Ilaria Liccardi, Julius Adebayo, Michael Muelly","submitted_at":"2020-11-10T22:23:25Z","abstract_excerpt":"We investigate whether post-hoc model explanations are effective for diagnosing model errors--model debugging. In response to the challenge of explaining a model's prediction, a vast array of explanation methods have been proposed. Despite increasing use, it is unclear if they are effective. To start, we categorize \\textit{bugs}, based on their source, into:~\\textit{data, model, and test-time} contamination bugs. For several explanation methods, we assess their ability to: detect spurious correlation artifacts (data contamination), diagnose mislabeled training examples (data contamination), di"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2011.05429","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/2011.05429/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":"2011.05429","created_at":"2026-07-05T01:50:49.879921+00:00"},{"alias_kind":"arxiv_version","alias_value":"2011.05429v1","created_at":"2026-07-05T01:50:49.879921+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2011.05429","created_at":"2026-07-05T01:50:49.879921+00:00"},{"alias_kind":"pith_short_12","alias_value":"R4625KAK2RSP","created_at":"2026-07-05T01:50:49.879921+00:00"},{"alias_kind":"pith_short_16","alias_value":"R4625KAK2RSPBL4D","created_at":"2026-07-05T01:50:49.879921+00:00"},{"alias_kind":"pith_short_8","alias_value":"R4625KAK","created_at":"2026-07-05T01:50:49.879921+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.20620","citing_title":"Dynamic Shapley Computation","ref_index":1,"is_internal_anchor":false},{"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":255,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16905","citing_title":"AIM: Adversarial Information Masking for Faithfulness Evaluation of Saliency Maps","ref_index":7,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/R4625KAK2RSPBL4DD4BCGMDQ3D","json":"https://pith.science/pith/R4625KAK2RSPBL4DD4BCGMDQ3D.json","graph_json":"https://pith.science/api/pith-number/R4625KAK2RSPBL4DD4BCGMDQ3D/graph.json","events_json":"https://pith.science/api/pith-number/R4625KAK2RSPBL4DD4BCGMDQ3D/events.json","paper":"https://pith.science/paper/R4625KAK"},"agent_actions":{"view_html":"https://pith.science/pith/R4625KAK2RSPBL4DD4BCGMDQ3D","download_json":"https://pith.science/pith/R4625KAK2RSPBL4DD4BCGMDQ3D.json","view_paper":"https://pith.science/paper/R4625KAK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2011.05429&json=true","fetch_graph":"https://pith.science/api/pith-number/R4625KAK2RSPBL4DD4BCGMDQ3D/graph.json","fetch_events":"https://pith.science/api/pith-number/R4625KAK2RSPBL4DD4BCGMDQ3D/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/R4625KAK2RSPBL4DD4BCGMDQ3D/action/timestamp_anchor","attest_storage":"https://pith.science/pith/R4625KAK2RSPBL4DD4BCGMDQ3D/action/storage_attestation","attest_author":"https://pith.science/pith/R4625KAK2RSPBL4DD4BCGMDQ3D/action/author_attestation","sign_citation":"https://pith.science/pith/R4625KAK2RSPBL4DD4BCGMDQ3D/action/citation_signature","submit_replication":"https://pith.science/pith/R4625KAK2RSPBL4DD4BCGMDQ3D/action/replication_record"}},"created_at":"2026-07-05T01:50:49.879921+00:00","updated_at":"2026-07-05T01:50:49.879921+00:00"}