{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:VJROJVCAP5IJ4DP7FSKK3V7TWG","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"1da4dc6dbfeede94edf8e456d6f83444bb59450e15d07e378e4cd705fc686d93","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-05T05:20:32Z","title_canon_sha256":"d489cfee78c064a58b81bc79d692ea47eaa40a80a8c91055d20724528d142adb"},"schema_version":"1.0","source":{"id":"2606.09899","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.09899","created_at":"2026-06-10T00:08:32Z"},{"alias_kind":"arxiv_version","alias_value":"2606.09899v1","created_at":"2026-06-10T00:08:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.09899","created_at":"2026-06-10T00:08:32Z"},{"alias_kind":"pith_short_12","alias_value":"VJROJVCAP5IJ","created_at":"2026-06-10T00:08:32Z"},{"alias_kind":"pith_short_16","alias_value":"VJROJVCAP5IJ4DP7","created_at":"2026-06-10T00:08:32Z"},{"alias_kind":"pith_short_8","alias_value":"VJROJVCA","created_at":"2026-06-10T00:08:32Z"}],"graph_snapshots":[{"event_id":"sha256:55694096feac4862520f9e224b1c8212f049c2bbd638d2dc5832a860eb8985aa","target":"graph","created_at":"2026-06-10T00:08:32Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2606.09899/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"A central goal of mechanistic interpretability is to identify which internal components causally drive a language model's behavior. Because these importance estimates serve as the evidence for identifying circuits, systematic errors can lead to the misidentification of the underlying mechanisms. While activation patching provides a gold-standard causal metric, its computational cost is prohibitive at scale. Practitioners instead rely on attribution patching, a gradient-based, first-order approximation whose reliability remains poorly understood. In this work, we characterize the source of this","authors_text":"Jialu Wang, Luyang Zhang","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-05T05:20:32Z","title":"When Attribution Patching Lies: Diagnosis and a Second-Order Correction"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.09899","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:131adaa9320bf97b0789005a67702df28db4971b502d71a2486a5e2aafe63941","target":"record","created_at":"2026-06-10T00:08:32Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"1da4dc6dbfeede94edf8e456d6f83444bb59450e15d07e378e4cd705fc686d93","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-05T05:20:32Z","title_canon_sha256":"d489cfee78c064a58b81bc79d692ea47eaa40a80a8c91055d20724528d142adb"},"schema_version":"1.0","source":{"id":"2606.09899","kind":"arxiv","version":1}},"canonical_sha256":"aa62e4d4407f509e0dff2c94add7f3b1976ebb7d25e2a15f187fd5467d38521f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"aa62e4d4407f509e0dff2c94add7f3b1976ebb7d25e2a15f187fd5467d38521f","first_computed_at":"2026-06-10T00:08:32.068903Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-10T00:08:32.068903Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"XGObt2Y+ev/dT05T+9vEJr6sbmKO5Eh/dPP0fokONg7E+i0eQBA4VZGoRHKnONR9qQDHnb9nBqqR56WlsGU6CA==","signature_status":"signed_v1","signed_at":"2026-06-10T00:08:32.069686Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.09899","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:131adaa9320bf97b0789005a67702df28db4971b502d71a2486a5e2aafe63941","sha256:55694096feac4862520f9e224b1c8212f049c2bbd638d2dc5832a860eb8985aa"],"state_sha256":"26fe7ce1700dd1dcd3d5bab09f674527936d99f987e22a1eacaf6b7c366058c1"}