{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:ACWEZZMSA7AQDHAT2OGROVTF7H","short_pith_number":"pith:ACWEZZMS","schema_version":"1.0","canonical_sha256":"00ac4ce59207c1019c13d38d175665f9c848436cc61609a5afba8bfcb31769ee","source":{"kind":"arxiv","id":"2109.08237","version":3},"attestation_state":"computed","paper":{"title":"Subtle Data Crimes: Naively training machine learning algorithms could lead to overly-optimistic results","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Efrat Shimron, Jonathan I. Tamir, Ke Wang, Michael Lustig","submitted_at":"2021-09-16T22:00:15Z","abstract_excerpt":"While open databases are an important resource in the Deep Learning (DL) era, they are sometimes used \"off-label\": data published for one task are used for training algorithms for a different one. This work aims to highlight that in some cases, this common practice may lead to biased, overly-optimistic results. We demonstrate this phenomenon for inverse problem solvers and show how their biased performance stems from hidden data preprocessing pipelines. We describe two preprocessing pipelines typical of open-access databases and study their effects on three well-established algorithms develope"},"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":"2109.08237","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-09-16T22:00:15Z","cross_cats_sorted":[],"title_canon_sha256":"e2b64364436f90d2b583a889720d5e43b7268e7fcf429969999ffb85ae5d01d8","abstract_canon_sha256":"c1d838304af79ce8b6ad48f189e1d5e0baa559b2898a0cb5620b10142c7e7cb7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:40:46.408193Z","signature_b64":"4HxWNqEB3QcQ3glCB29ePoyojHAgiOLmOo1DW5uMyJWYE6VMn8m0Jp9LDlvz4UBvvbim7YTrb0rxuM4hrs0uDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"00ac4ce59207c1019c13d38d175665f9c848436cc61609a5afba8bfcb31769ee","last_reissued_at":"2026-07-05T03:40:46.407697Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:40:46.407697Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Subtle Data Crimes: Naively training machine learning algorithms could lead to overly-optimistic results","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Efrat Shimron, Jonathan I. Tamir, Ke Wang, Michael Lustig","submitted_at":"2021-09-16T22:00:15Z","abstract_excerpt":"While open databases are an important resource in the Deep Learning (DL) era, they are sometimes used \"off-label\": data published for one task are used for training algorithms for a different one. This work aims to highlight that in some cases, this common practice may lead to biased, overly-optimistic results. We demonstrate this phenomenon for inverse problem solvers and show how their biased performance stems from hidden data preprocessing pipelines. We describe two preprocessing pipelines typical of open-access databases and study their effects on three well-established algorithms develope"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2109.08237","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/2109.08237/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":"2109.08237","created_at":"2026-07-05T03:40:46.407768+00:00"},{"alias_kind":"arxiv_version","alias_value":"2109.08237v3","created_at":"2026-07-05T03:40:46.407768+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2109.08237","created_at":"2026-07-05T03:40:46.407768+00:00"},{"alias_kind":"pith_short_12","alias_value":"ACWEZZMSA7AQ","created_at":"2026-07-05T03:40:46.407768+00:00"},{"alias_kind":"pith_short_16","alias_value":"ACWEZZMSA7AQDHAT","created_at":"2026-07-05T03:40:46.407768+00:00"},{"alias_kind":"pith_short_8","alias_value":"ACWEZZMS","created_at":"2026-07-05T03:40:46.407768+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/ACWEZZMSA7AQDHAT2OGROVTF7H","json":"https://pith.science/pith/ACWEZZMSA7AQDHAT2OGROVTF7H.json","graph_json":"https://pith.science/api/pith-number/ACWEZZMSA7AQDHAT2OGROVTF7H/graph.json","events_json":"https://pith.science/api/pith-number/ACWEZZMSA7AQDHAT2OGROVTF7H/events.json","paper":"https://pith.science/paper/ACWEZZMS"},"agent_actions":{"view_html":"https://pith.science/pith/ACWEZZMSA7AQDHAT2OGROVTF7H","download_json":"https://pith.science/pith/ACWEZZMSA7AQDHAT2OGROVTF7H.json","view_paper":"https://pith.science/paper/ACWEZZMS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2109.08237&json=true","fetch_graph":"https://pith.science/api/pith-number/ACWEZZMSA7AQDHAT2OGROVTF7H/graph.json","fetch_events":"https://pith.science/api/pith-number/ACWEZZMSA7AQDHAT2OGROVTF7H/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ACWEZZMSA7AQDHAT2OGROVTF7H/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ACWEZZMSA7AQDHAT2OGROVTF7H/action/storage_attestation","attest_author":"https://pith.science/pith/ACWEZZMSA7AQDHAT2OGROVTF7H/action/author_attestation","sign_citation":"https://pith.science/pith/ACWEZZMSA7AQDHAT2OGROVTF7H/action/citation_signature","submit_replication":"https://pith.science/pith/ACWEZZMSA7AQDHAT2OGROVTF7H/action/replication_record"}},"created_at":"2026-07-05T03:40:46.407768+00:00","updated_at":"2026-07-05T03:40:46.407768+00:00"}