{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:2I7VWK2LEK6JT5BNZWUBJYIGOH","short_pith_number":"pith:2I7VWK2L","schema_version":"1.0","canonical_sha256":"d23f5b2b4b22bc99f42dcda814e10671e0fb1892fcd524f73da32bee75c8b9b8","source":{"kind":"arxiv","id":"2401.06654","version":1},"attestation_state":"computed","paper":{"title":"Decoupling Pixel Flipping and Occlusion Strategy for Consistent XAI Benchmarks","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Johanna Vielhaben, Nils Strodthoff, Stefan Bl\\\"ucher","submitted_at":"2024-01-12T16:01:17Z","abstract_excerpt":"Feature removal is a central building block for eXplainable AI (XAI), both for occlusion-based explanations (Shapley values) as well as their evaluation (pixel flipping, PF). However, occlusion strategies can vary significantly from simple mean replacement up to inpainting with state-of-the-art diffusion models. This ambiguity limits the usefulness of occlusion-based approaches. For example, PF benchmarks lead to contradicting rankings. This is amplified by competing PF measures: Features are either removed starting with most influential first (MIF) or least influential first (LIF). This study"},"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":"2401.06654","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CV","submitted_at":"2024-01-12T16:01:17Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"0ecd904d2c144e3c1844bbae248e2113f01e2ba5152c6dce1e44f1bb4e63c297","abstract_canon_sha256":"04a2fa5925e48cc19d60e2588e3b0dd5c86f4a4ee17a24b01131b9bbc1a0c393"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:59:05.201488Z","signature_b64":"nd1TTNFalMkhbxRneVpZQA6DPRSXLd+kQ2QCB8t/KzQ+opY3llwottNszxFMXign++5B+SVfg92YmBiGSRMWAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d23f5b2b4b22bc99f42dcda814e10671e0fb1892fcd524f73da32bee75c8b9b8","last_reissued_at":"2026-07-05T08:59:05.201058Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:59:05.201058Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Decoupling Pixel Flipping and Occlusion Strategy for Consistent XAI Benchmarks","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Johanna Vielhaben, Nils Strodthoff, Stefan Bl\\\"ucher","submitted_at":"2024-01-12T16:01:17Z","abstract_excerpt":"Feature removal is a central building block for eXplainable AI (XAI), both for occlusion-based explanations (Shapley values) as well as their evaluation (pixel flipping, PF). However, occlusion strategies can vary significantly from simple mean replacement up to inpainting with state-of-the-art diffusion models. This ambiguity limits the usefulness of occlusion-based approaches. For example, PF benchmarks lead to contradicting rankings. This is amplified by competing PF measures: Features are either removed starting with most influential first (MIF) or least influential first (LIF). This study"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.06654","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/2401.06654/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":"2401.06654","created_at":"2026-07-05T08:59:05.201118+00:00"},{"alias_kind":"arxiv_version","alias_value":"2401.06654v1","created_at":"2026-07-05T08:59:05.201118+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.06654","created_at":"2026-07-05T08:59:05.201118+00:00"},{"alias_kind":"pith_short_12","alias_value":"2I7VWK2LEK6J","created_at":"2026-07-05T08:59:05.201118+00:00"},{"alias_kind":"pith_short_16","alias_value":"2I7VWK2LEK6JT5BN","created_at":"2026-07-05T08:59:05.201118+00:00"},{"alias_kind":"pith_short_8","alias_value":"2I7VWK2L","created_at":"2026-07-05T08:59:05.201118+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.12203","citing_title":"Explainability for Vision Foundation Models: A Survey","ref_index":150,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2I7VWK2LEK6JT5BNZWUBJYIGOH","json":"https://pith.science/pith/2I7VWK2LEK6JT5BNZWUBJYIGOH.json","graph_json":"https://pith.science/api/pith-number/2I7VWK2LEK6JT5BNZWUBJYIGOH/graph.json","events_json":"https://pith.science/api/pith-number/2I7VWK2LEK6JT5BNZWUBJYIGOH/events.json","paper":"https://pith.science/paper/2I7VWK2L"},"agent_actions":{"view_html":"https://pith.science/pith/2I7VWK2LEK6JT5BNZWUBJYIGOH","download_json":"https://pith.science/pith/2I7VWK2LEK6JT5BNZWUBJYIGOH.json","view_paper":"https://pith.science/paper/2I7VWK2L","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2401.06654&json=true","fetch_graph":"https://pith.science/api/pith-number/2I7VWK2LEK6JT5BNZWUBJYIGOH/graph.json","fetch_events":"https://pith.science/api/pith-number/2I7VWK2LEK6JT5BNZWUBJYIGOH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2I7VWK2LEK6JT5BNZWUBJYIGOH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2I7VWK2LEK6JT5BNZWUBJYIGOH/action/storage_attestation","attest_author":"https://pith.science/pith/2I7VWK2LEK6JT5BNZWUBJYIGOH/action/author_attestation","sign_citation":"https://pith.science/pith/2I7VWK2LEK6JT5BNZWUBJYIGOH/action/citation_signature","submit_replication":"https://pith.science/pith/2I7VWK2LEK6JT5BNZWUBJYIGOH/action/replication_record"}},"created_at":"2026-07-05T08:59:05.201118+00:00","updated_at":"2026-07-05T08:59:05.201118+00:00"}