{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:UVW5XR22A5PKMF4OITZ5KUBCVA","short_pith_number":"pith:UVW5XR22","schema_version":"1.0","canonical_sha256":"a56ddbc75a075ea6178e44f3d55022a82d945234b111ad05c8fde80efdaf281d","source":{"kind":"arxiv","id":"2401.08513","version":3},"attestation_state":"computed","paper":{"title":"X Hacking: The Threat of Misguided AutoML","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CR"],"primary_cat":"cs.LG","authors_text":"Andrea \\v{S}ipka, David Selby, Eyke H\\\"ullermeier, Rahul Sharma, Sebastian Vollmer, Sergey Redyuk, Sumantrak Mukherjee","submitted_at":"2024-01-16T17:21:33Z","abstract_excerpt":"Explainable AI (XAI) and interpretable machine learning methods help to build trust in model predictions and derived insights, yet also present a perverse incentive for analysts to manipulate XAI metrics to support pre-specified conclusions. This paper introduces the concept of X-hacking, a form of p-hacking applied to XAI metrics such as SHAP values. We show how easily an automated machine learning pipeline can be adapted to exploit model multiplicity at scale: searching a Rashomon set of 'defensible' models with similar predictive performance to find a desired explanation. We formulate the t"},"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":"2401.08513","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-01-16T17:21:33Z","cross_cats_sorted":["cs.CR"],"title_canon_sha256":"8f4b5060f8e37b03bd49b8c38367dadbabdf77c96225206d11ab8070ac3be625","abstract_canon_sha256":"36b25481162f0096cdb0da4eb2f830a6a33b33a66af6b8fb74f5fb832e5af167"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:37:08.029170Z","signature_b64":"i+EkR/RCfIpA2xqX8/44KRpsFI4iywuOipmPOGZPjFvM1fVGlwmqNBk7DEPlwc/RqTtvNGSvj37svcrWlHBsAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a56ddbc75a075ea6178e44f3d55022a82d945234b111ad05c8fde80efdaf281d","last_reissued_at":"2026-07-05T11:37:08.028729Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:37:08.028729Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"X Hacking: The Threat of Misguided AutoML","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CR"],"primary_cat":"cs.LG","authors_text":"Andrea \\v{S}ipka, David Selby, Eyke H\\\"ullermeier, Rahul Sharma, Sebastian Vollmer, Sergey Redyuk, Sumantrak Mukherjee","submitted_at":"2024-01-16T17:21:33Z","abstract_excerpt":"Explainable AI (XAI) and interpretable machine learning methods help to build trust in model predictions and derived insights, yet also present a perverse incentive for analysts to manipulate XAI metrics to support pre-specified conclusions. This paper introduces the concept of X-hacking, a form of p-hacking applied to XAI metrics such as SHAP values. We show how easily an automated machine learning pipeline can be adapted to exploit model multiplicity at scale: searching a Rashomon set of 'defensible' models with similar predictive performance to find a desired explanation. We formulate the t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.08513","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/2401.08513/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":"2401.08513","created_at":"2026-07-05T11:37:08.028787+00:00"},{"alias_kind":"arxiv_version","alias_value":"2401.08513v3","created_at":"2026-07-05T11:37:08.028787+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.08513","created_at":"2026-07-05T11:37:08.028787+00:00"},{"alias_kind":"pith_short_12","alias_value":"UVW5XR22A5PK","created_at":"2026-07-05T11:37:08.028787+00:00"},{"alias_kind":"pith_short_16","alias_value":"UVW5XR22A5PKMF4O","created_at":"2026-07-05T11:37:08.028787+00:00"},{"alias_kind":"pith_short_8","alias_value":"UVW5XR22","created_at":"2026-07-05T11:37:08.028787+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2502.03946","citing_title":"CleanSurvival: Automated data preprocessing for time-to-event models using reinforcement learning","ref_index":3,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UVW5XR22A5PKMF4OITZ5KUBCVA","json":"https://pith.science/pith/UVW5XR22A5PKMF4OITZ5KUBCVA.json","graph_json":"https://pith.science/api/pith-number/UVW5XR22A5PKMF4OITZ5KUBCVA/graph.json","events_json":"https://pith.science/api/pith-number/UVW5XR22A5PKMF4OITZ5KUBCVA/events.json","paper":"https://pith.science/paper/UVW5XR22"},"agent_actions":{"view_html":"https://pith.science/pith/UVW5XR22A5PKMF4OITZ5KUBCVA","download_json":"https://pith.science/pith/UVW5XR22A5PKMF4OITZ5KUBCVA.json","view_paper":"https://pith.science/paper/UVW5XR22","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2401.08513&json=true","fetch_graph":"https://pith.science/api/pith-number/UVW5XR22A5PKMF4OITZ5KUBCVA/graph.json","fetch_events":"https://pith.science/api/pith-number/UVW5XR22A5PKMF4OITZ5KUBCVA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UVW5XR22A5PKMF4OITZ5KUBCVA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UVW5XR22A5PKMF4OITZ5KUBCVA/action/storage_attestation","attest_author":"https://pith.science/pith/UVW5XR22A5PKMF4OITZ5KUBCVA/action/author_attestation","sign_citation":"https://pith.science/pith/UVW5XR22A5PKMF4OITZ5KUBCVA/action/citation_signature","submit_replication":"https://pith.science/pith/UVW5XR22A5PKMF4OITZ5KUBCVA/action/replication_record"}},"created_at":"2026-07-05T11:37:08.028787+00:00","updated_at":"2026-07-05T11:37:08.028787+00:00"}