{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:4NOURMLEX3ZYLYDGRZQDELAT4P","short_pith_number":"pith:4NOURMLE","schema_version":"1.0","canonical_sha256":"e35d48b164bef385e0668e60322c13e3d07e582d30c68f1f274a517049270a03","source":{"kind":"arxiv","id":"2407.04690","version":1},"attestation_state":"computed","paper":{"title":"Missed Causes and Ambiguous Effects: Counterfactuals Pose Challenges for Interpreting Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Aaron Mueller","submitted_at":"2024-07-05T17:53:03Z","abstract_excerpt":"Interpretability research takes counterfactual theories of causality for granted. Most causal methods rely on counterfactual interventions to inputs or the activations of particular model components, followed by observations of the change in models' output logits or behaviors. While this yields more faithful evidence than correlational methods, counterfactuals nonetheless have key problems that bias our findings in specific and predictable ways. Specifically, (i) counterfactual theories do not effectively capture multiple independently sufficient causes of the same effect, which leads us to mi"},"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":"2407.04690","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-07-05T17:53:03Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"fd8e57058ec8e81ed2efcf1da72f3d006c0421848bc0b09175d2ee13c42386f1","abstract_canon_sha256":"80faa4daf902234bf688de23ea3c4dfe7218715ddc00e5e081d94b1c0e77a55c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:40:37.495262Z","signature_b64":"CFxvCIOYjjTamiR29iVxryqNJnteGvlqK+A0Cxhb/6q1CJggYgJmGK1V64KEr+rxON+yPqkoR6FUAREioRgwBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e35d48b164bef385e0668e60322c13e3d07e582d30c68f1f274a517049270a03","last_reissued_at":"2026-07-05T08:40:37.494841Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:40:37.494841Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Missed Causes and Ambiguous Effects: Counterfactuals Pose Challenges for Interpreting Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Aaron Mueller","submitted_at":"2024-07-05T17:53:03Z","abstract_excerpt":"Interpretability research takes counterfactual theories of causality for granted. Most causal methods rely on counterfactual interventions to inputs or the activations of particular model components, followed by observations of the change in models' output logits or behaviors. While this yields more faithful evidence than correlational methods, counterfactuals nonetheless have key problems that bias our findings in specific and predictable ways. Specifically, (i) counterfactual theories do not effectively capture multiple independently sufficient causes of the same effect, which leads us to mi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.04690","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/2407.04690/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":"2407.04690","created_at":"2026-07-05T08:40:37.494894+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.04690v1","created_at":"2026-07-05T08:40:37.494894+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.04690","created_at":"2026-07-05T08:40:37.494894+00:00"},{"alias_kind":"pith_short_12","alias_value":"4NOURMLEX3ZY","created_at":"2026-07-05T08:40:37.494894+00:00"},{"alias_kind":"pith_short_16","alias_value":"4NOURMLEX3ZYLYDG","created_at":"2026-07-05T08:40:37.494894+00:00"},{"alias_kind":"pith_short_8","alias_value":"4NOURMLE","created_at":"2026-07-05T08:40:37.494894+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.25092","citing_title":"How Modular Is a Frontier Mixture-of-Experts? A Pre-registered Causal Test in Which Apparent Expert Modularity Mostly Dissolves","ref_index":8,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4NOURMLEX3ZYLYDGRZQDELAT4P","json":"https://pith.science/pith/4NOURMLEX3ZYLYDGRZQDELAT4P.json","graph_json":"https://pith.science/api/pith-number/4NOURMLEX3ZYLYDGRZQDELAT4P/graph.json","events_json":"https://pith.science/api/pith-number/4NOURMLEX3ZYLYDGRZQDELAT4P/events.json","paper":"https://pith.science/paper/4NOURMLE"},"agent_actions":{"view_html":"https://pith.science/pith/4NOURMLEX3ZYLYDGRZQDELAT4P","download_json":"https://pith.science/pith/4NOURMLEX3ZYLYDGRZQDELAT4P.json","view_paper":"https://pith.science/paper/4NOURMLE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.04690&json=true","fetch_graph":"https://pith.science/api/pith-number/4NOURMLEX3ZYLYDGRZQDELAT4P/graph.json","fetch_events":"https://pith.science/api/pith-number/4NOURMLEX3ZYLYDGRZQDELAT4P/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4NOURMLEX3ZYLYDGRZQDELAT4P/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4NOURMLEX3ZYLYDGRZQDELAT4P/action/storage_attestation","attest_author":"https://pith.science/pith/4NOURMLEX3ZYLYDGRZQDELAT4P/action/author_attestation","sign_citation":"https://pith.science/pith/4NOURMLEX3ZYLYDGRZQDELAT4P/action/citation_signature","submit_replication":"https://pith.science/pith/4NOURMLEX3ZYLYDGRZQDELAT4P/action/replication_record"}},"created_at":"2026-07-05T08:40:37.494894+00:00","updated_at":"2026-07-05T08:40:37.494894+00:00"}