{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:CVGV3CCRJL4Y42DJ43RLY7ISHU","short_pith_number":"pith:CVGV3CCR","schema_version":"1.0","canonical_sha256":"154d5d88514af98e6869e6e2bc7d123d10f8ec46c3c82ea0503555cc5c4943a6","source":{"kind":"arxiv","id":"2207.07048","version":1},"attestation_state":"computed","paper":{"title":"Leakage and the Reproducibility Crisis in ML-based Science","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ME"],"primary_cat":"cs.LG","authors_text":"Arvind Narayanan, Sayash Kapoor","submitted_at":"2022-07-14T16:44:59Z","abstract_excerpt":"The use of machine learning (ML) methods for prediction and forecasting has become widespread across the quantitative sciences. However, there are many known methodological pitfalls, including data leakage, in ML-based science. In this paper, we systematically investigate reproducibility issues in ML-based science. We show that data leakage is indeed a widespread problem and has led to severe reproducibility failures. Specifically, through a survey of literature in research communities that adopted ML methods, we find 17 fields where errors have been found, collectively affecting 329 papers an"},"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":"2207.07048","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-07-14T16:44:59Z","cross_cats_sorted":["cs.AI","stat.ME"],"title_canon_sha256":"529a3f2c0740807242d2c0e4cb49ed18c2f76f030878313ac98972c9b57c9964","abstract_canon_sha256":"3d06f00faf5243a0b4ab803d026c8e8da1e5e011df7ef382dce784e58e24b7bf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:40:20.375618Z","signature_b64":"HJT+tATFRdsMMLsZT1XtEX8V1NHUQQSAnKn5AkMDdZi4mf35CvJGsHmwk4c9DEq2BvZR+kISdE4vt5JksZPTDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"154d5d88514af98e6869e6e2bc7d123d10f8ec46c3c82ea0503555cc5c4943a6","last_reissued_at":"2026-07-05T04:40:20.375180Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:40:20.375180Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Leakage and the Reproducibility Crisis in ML-based Science","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ME"],"primary_cat":"cs.LG","authors_text":"Arvind Narayanan, Sayash Kapoor","submitted_at":"2022-07-14T16:44:59Z","abstract_excerpt":"The use of machine learning (ML) methods for prediction and forecasting has become widespread across the quantitative sciences. However, there are many known methodological pitfalls, including data leakage, in ML-based science. In this paper, we systematically investigate reproducibility issues in ML-based science. We show that data leakage is indeed a widespread problem and has led to severe reproducibility failures. Specifically, through a survey of literature in research communities that adopted ML methods, we find 17 fields where errors have been found, collectively affecting 329 papers an"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2207.07048","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/2207.07048/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":"2207.07048","created_at":"2026-07-05T04:40:20.375236+00:00"},{"alias_kind":"arxiv_version","alias_value":"2207.07048v1","created_at":"2026-07-05T04:40:20.375236+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2207.07048","created_at":"2026-07-05T04:40:20.375236+00:00"},{"alias_kind":"pith_short_12","alias_value":"CVGV3CCRJL4Y","created_at":"2026-07-05T04:40:20.375236+00:00"},{"alias_kind":"pith_short_16","alias_value":"CVGV3CCRJL4Y42DJ","created_at":"2026-07-05T04:40:20.375236+00:00"},{"alias_kind":"pith_short_8","alias_value":"CVGV3CCR","created_at":"2026-07-05T04:40:20.375236+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2607.06736","citing_title":"The Generalization Gap in Machine Learning EoS Inference from Core-Collapse Supernova Gravitational Waves","ref_index":28,"is_internal_anchor":true},{"citing_arxiv_id":"2607.05282","citing_title":"Wavelet Scattering Transform for Interpretable Schizophrenia Biomarker Discovery and Classification from Resting-State EEG","ref_index":27,"is_internal_anchor":true},{"citing_arxiv_id":"2606.22053","citing_title":"Gradient-Descent Steps to Success over Mean Accuracy: A Paradigm Shift for ML","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2606.18054","citing_title":"AI-based Cognitive-linguistic Features for Dementia Assessment in Picture Description","ref_index":51,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23628","citing_title":"How Hard is it to Rig a Benchmark? A Social Choice Analysis of Leaderboard Robustness","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13801","citing_title":"Improving Reproducibility in Evaluation through Multi-Level Annotator Modeling","ref_index":17,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CVGV3CCRJL4Y42DJ43RLY7ISHU","json":"https://pith.science/pith/CVGV3CCRJL4Y42DJ43RLY7ISHU.json","graph_json":"https://pith.science/api/pith-number/CVGV3CCRJL4Y42DJ43RLY7ISHU/graph.json","events_json":"https://pith.science/api/pith-number/CVGV3CCRJL4Y42DJ43RLY7ISHU/events.json","paper":"https://pith.science/paper/CVGV3CCR"},"agent_actions":{"view_html":"https://pith.science/pith/CVGV3CCRJL4Y42DJ43RLY7ISHU","download_json":"https://pith.science/pith/CVGV3CCRJL4Y42DJ43RLY7ISHU.json","view_paper":"https://pith.science/paper/CVGV3CCR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2207.07048&json=true","fetch_graph":"https://pith.science/api/pith-number/CVGV3CCRJL4Y42DJ43RLY7ISHU/graph.json","fetch_events":"https://pith.science/api/pith-number/CVGV3CCRJL4Y42DJ43RLY7ISHU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CVGV3CCRJL4Y42DJ43RLY7ISHU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CVGV3CCRJL4Y42DJ43RLY7ISHU/action/storage_attestation","attest_author":"https://pith.science/pith/CVGV3CCRJL4Y42DJ43RLY7ISHU/action/author_attestation","sign_citation":"https://pith.science/pith/CVGV3CCRJL4Y42DJ43RLY7ISHU/action/citation_signature","submit_replication":"https://pith.science/pith/CVGV3CCRJL4Y42DJ43RLY7ISHU/action/replication_record"}},"created_at":"2026-07-05T04:40:20.375236+00:00","updated_at":"2026-07-05T04:40:20.375236+00:00"}