{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:6LFWVNPK7SX3Q5G22A6FTCAFTZ","short_pith_number":"pith:6LFWVNPK","schema_version":"1.0","canonical_sha256":"f2cb6ab5eafcafb874dad03c5988059e46b20078c3485a55cbd80780838ef5d9","source":{"kind":"arxiv","id":"2608.00566","version":1},"attestation_state":"computed","paper":{"title":"Crushing the Evidence: A Dual-Penalty Evasion Framework for Fooling White-Box Explainable AI Auditors","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Harsh Kasyap, Niraj Kumar","submitted_at":"2026-08-01T10:03:14Z","abstract_excerpt":"Post-hoc model explainers such as LIME, SHAP, and Integrated Gradients are widely deployed to audit models in high-stakes sensitive domains, including finance, healthcare, and social welfare. This ensures the model's transparency and acceptability. However, a few studies have examined potential attacks in the explainability pipeline. Adversaries can attempt to conceal algorithmic biases or backdoors using adversarial explanation attacks. These attacks have relied on scaffolding out-of-distribution (OOD) detectors that toggle predictions when queried by an explainer. Consequently, defenses have"},"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":"2608.00566","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-08-01T10:03:14Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"7e96f346b65c50fbc9b39f147d5c398f97a841c820eca6f15aa58cef30c27a97","abstract_canon_sha256":"89d89f3c5f1d09c6fd7d560fc7af78dfd6382e1c954c7af855a303222b7c2b51"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-04T00:37:26.935494Z","signature_b64":"qakB+6/JTvC7aNykwF8qwUG4chfRDEYRCvKhGrb6qH/tG9JwylKiWDQzaCPFQT4DLfy59DsHj8jZ1/U7S43rBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f2cb6ab5eafcafb874dad03c5988059e46b20078c3485a55cbd80780838ef5d9","last_reissued_at":"2026-08-04T00:37:26.933966Z","signature_status":"signed_v1","first_computed_at":"2026-08-04T00:37:26.933966Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Crushing the Evidence: A Dual-Penalty Evasion Framework for Fooling White-Box Explainable AI Auditors","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Harsh Kasyap, Niraj Kumar","submitted_at":"2026-08-01T10:03:14Z","abstract_excerpt":"Post-hoc model explainers such as LIME, SHAP, and Integrated Gradients are widely deployed to audit models in high-stakes sensitive domains, including finance, healthcare, and social welfare. This ensures the model's transparency and acceptability. However, a few studies have examined potential attacks in the explainability pipeline. Adversaries can attempt to conceal algorithmic biases or backdoors using adversarial explanation attacks. These attacks have relied on scaffolding out-of-distribution (OOD) detectors that toggle predictions when queried by an explainer. Consequently, defenses have"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.00566","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/2608.00566/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":"2608.00566","created_at":"2026-08-04T00:37:26.935241+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.00566v1","created_at":"2026-08-04T00:37:26.935241+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.00566","created_at":"2026-08-04T00:37:26.935241+00:00"},{"alias_kind":"pith_short_12","alias_value":"6LFWVNPK7SX3","created_at":"2026-08-04T00:37:26.935241+00:00"},{"alias_kind":"pith_short_16","alias_value":"6LFWVNPK7SX3Q5G2","created_at":"2026-08-04T00:37:26.935241+00:00"},{"alias_kind":"pith_short_8","alias_value":"6LFWVNPK","created_at":"2026-08-04T00:37:26.935241+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/6LFWVNPK7SX3Q5G22A6FTCAFTZ","json":"https://pith.science/pith/6LFWVNPK7SX3Q5G22A6FTCAFTZ.json","graph_json":"https://pith.science/api/pith-number/6LFWVNPK7SX3Q5G22A6FTCAFTZ/graph.json","events_json":"https://pith.science/api/pith-number/6LFWVNPK7SX3Q5G22A6FTCAFTZ/events.json","paper":"https://pith.science/paper/6LFWVNPK"},"agent_actions":{"view_html":"https://pith.science/pith/6LFWVNPK7SX3Q5G22A6FTCAFTZ","download_json":"https://pith.science/pith/6LFWVNPK7SX3Q5G22A6FTCAFTZ.json","view_paper":"https://pith.science/paper/6LFWVNPK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.00566&json=true","fetch_graph":"https://pith.science/api/pith-number/6LFWVNPK7SX3Q5G22A6FTCAFTZ/graph.json","fetch_events":"https://pith.science/api/pith-number/6LFWVNPK7SX3Q5G22A6FTCAFTZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6LFWVNPK7SX3Q5G22A6FTCAFTZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6LFWVNPK7SX3Q5G22A6FTCAFTZ/action/storage_attestation","attest_author":"https://pith.science/pith/6LFWVNPK7SX3Q5G22A6FTCAFTZ/action/author_attestation","sign_citation":"https://pith.science/pith/6LFWVNPK7SX3Q5G22A6FTCAFTZ/action/citation_signature","submit_replication":"https://pith.science/pith/6LFWVNPK7SX3Q5G22A6FTCAFTZ/action/replication_record"}},"created_at":"2026-08-04T00:37:26.935241+00:00","updated_at":"2026-08-04T00:37:26.935241+00:00"}