{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:UQJ4RJ2KGAZY6PXUQF5CP46ZD3","short_pith_number":"pith:UQJ4RJ2K","schema_version":"1.0","canonical_sha256":"a413c8a74a30338f3ef4817a27f3d91edec3ea7355720b3315f5f619cbca386e","source":{"kind":"arxiv","id":"2607.25546","version":1},"attestation_state":"computed","paper":{"title":"From Training to Deployment: Post-Hoc Causal Feature Identification via Sensitivity Ratios","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Athanasios Vlontzos, Bernhard Kainz, Giorgos Papanastasiou, Sotirios Tsaftaris","submitted_at":"2026-07-28T10:26:09Z","abstract_excerpt":"Given a model that is already trained, which features does it rely on causally versus spuriously? Existing methods require access to the training procedure and cannot answer this post-hoc. We introduce the \\textbf{Normalised Sensitivity Ratio~(NSR)}, a post-hoc, model-agnostic diagnostic for this question under a structured-shift regime: environments differ primarily in the mean of spurious features while the causal mechanism and causal marginals remain stable, as in multi-site clinical data or multi-batch genomics. Within this regime, causal features induce constant model sensitivity across e"},"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":"2607.25546","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-28T10:26:09Z","cross_cats_sorted":[],"title_canon_sha256":"b9442b701a8deb1a5c45837bdced0ce357df5e37fb683c1e4f36ff783ce01483","abstract_canon_sha256":"47d7ecdaa6b7ff8b4500e7274501d3b36c778c2574fbb7b4f6e651709dd216a6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-29T01:25:46.207759Z","signature_b64":"G4Y90+csRCkqlKhu7fT98B1T/wo2yL4YSNJ4TCS/LarwgU/haX4Yh3uFLXwiQ9zK/uIgf6+YmnligIe76foGCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a413c8a74a30338f3ef4817a27f3d91edec3ea7355720b3315f5f619cbca386e","last_reissued_at":"2026-07-29T01:25:46.206943Z","signature_status":"signed_v1","first_computed_at":"2026-07-29T01:25:46.206943Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"From Training to Deployment: Post-Hoc Causal Feature Identification via Sensitivity Ratios","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Athanasios Vlontzos, Bernhard Kainz, Giorgos Papanastasiou, Sotirios Tsaftaris","submitted_at":"2026-07-28T10:26:09Z","abstract_excerpt":"Given a model that is already trained, which features does it rely on causally versus spuriously? Existing methods require access to the training procedure and cannot answer this post-hoc. We introduce the \\textbf{Normalised Sensitivity Ratio~(NSR)}, a post-hoc, model-agnostic diagnostic for this question under a structured-shift regime: environments differ primarily in the mean of spurious features while the causal mechanism and causal marginals remain stable, as in multi-site clinical data or multi-batch genomics. Within this regime, causal features induce constant model sensitivity across e"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.25546","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/2607.25546/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":"2607.25546","created_at":"2026-07-29T01:25:46.207365+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.25546v1","created_at":"2026-07-29T01:25:46.207365+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.25546","created_at":"2026-07-29T01:25:46.207365+00:00"},{"alias_kind":"pith_short_12","alias_value":"UQJ4RJ2KGAZY","created_at":"2026-07-29T01:25:46.207365+00:00"},{"alias_kind":"pith_short_16","alias_value":"UQJ4RJ2KGAZY6PXU","created_at":"2026-07-29T01:25:46.207365+00:00"},{"alias_kind":"pith_short_8","alias_value":"UQJ4RJ2K","created_at":"2026-07-29T01:25:46.207365+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/UQJ4RJ2KGAZY6PXUQF5CP46ZD3","json":"https://pith.science/pith/UQJ4RJ2KGAZY6PXUQF5CP46ZD3.json","graph_json":"https://pith.science/api/pith-number/UQJ4RJ2KGAZY6PXUQF5CP46ZD3/graph.json","events_json":"https://pith.science/api/pith-number/UQJ4RJ2KGAZY6PXUQF5CP46ZD3/events.json","paper":"https://pith.science/paper/UQJ4RJ2K"},"agent_actions":{"view_html":"https://pith.science/pith/UQJ4RJ2KGAZY6PXUQF5CP46ZD3","download_json":"https://pith.science/pith/UQJ4RJ2KGAZY6PXUQF5CP46ZD3.json","view_paper":"https://pith.science/paper/UQJ4RJ2K","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.25546&json=true","fetch_graph":"https://pith.science/api/pith-number/UQJ4RJ2KGAZY6PXUQF5CP46ZD3/graph.json","fetch_events":"https://pith.science/api/pith-number/UQJ4RJ2KGAZY6PXUQF5CP46ZD3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UQJ4RJ2KGAZY6PXUQF5CP46ZD3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UQJ4RJ2KGAZY6PXUQF5CP46ZD3/action/storage_attestation","attest_author":"https://pith.science/pith/UQJ4RJ2KGAZY6PXUQF5CP46ZD3/action/author_attestation","sign_citation":"https://pith.science/pith/UQJ4RJ2KGAZY6PXUQF5CP46ZD3/action/citation_signature","submit_replication":"https://pith.science/pith/UQJ4RJ2KGAZY6PXUQF5CP46ZD3/action/replication_record"}},"created_at":"2026-07-29T01:25:46.207365+00:00","updated_at":"2026-07-29T01:25:46.207365+00:00"}