{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:FO52FBCUDUET6OFTZ23JRUJA7G","short_pith_number":"pith:FO52FBCU","schema_version":"1.0","canonical_sha256":"2bbba284541d093f38b3ceb698d120f9b789b2069eea28ed706f555a06446d59","source":{"kind":"arxiv","id":"2605.27569","version":1},"attestation_state":"computed","paper":{"title":"RULER: Representation-Level Verification of Machine Unlearning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Axel Finke, Georgina Cosma","submitted_at":"2026-05-26T18:41:48Z","abstract_excerpt":"Machine unlearning aims to remove the influence of specific training records from a deployed model without retraining from scratch. Current protocols verify this at the output level through membership inference, retain accuracy, and forget-set accuracy, but a model can satisfy all three whilst still encoding forgotten records in its intermediate representations. We introduce RULER, a set of representation-level verification metrics. The oracle-comparative metric M2 measures whether forget-set records occupy the same representational position as in a model retrained without them. The oracle-fre"},"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":"2605.27569","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-26T18:41:48Z","cross_cats_sorted":[],"title_canon_sha256":"0446f472d26dc9c48c711c607ecbd13bf932ffb01af394884d22a7efc6d8a5fe","abstract_canon_sha256":"6b587c2ed946a5e11617c38bc50fd2599077862976268a7ed79d2a3a88513840"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-28T01:04:16.193883Z","signature_b64":"fslNqseUUmSRABSvLpyylbyN/vRKYO0+33SjW2J4b8Cp4+MYkixYc9hcnpoyWY+w4ctcjAhcCu5WVbcynEHgBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2bbba284541d093f38b3ceb698d120f9b789b2069eea28ed706f555a06446d59","last_reissued_at":"2026-05-28T01:04:16.193471Z","signature_status":"signed_v1","first_computed_at":"2026-05-28T01:04:16.193471Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RULER: Representation-Level Verification of Machine Unlearning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Axel Finke, Georgina Cosma","submitted_at":"2026-05-26T18:41:48Z","abstract_excerpt":"Machine unlearning aims to remove the influence of specific training records from a deployed model without retraining from scratch. Current protocols verify this at the output level through membership inference, retain accuracy, and forget-set accuracy, but a model can satisfy all three whilst still encoding forgotten records in its intermediate representations. We introduce RULER, a set of representation-level verification metrics. The oracle-comparative metric M2 measures whether forget-set records occupy the same representational position as in a model retrained without them. The oracle-fre"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.27569","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/2605.27569/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":"2605.27569","created_at":"2026-05-28T01:04:16.193533+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.27569v1","created_at":"2026-05-28T01:04:16.193533+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.27569","created_at":"2026-05-28T01:04:16.193533+00:00"},{"alias_kind":"pith_short_12","alias_value":"FO52FBCUDUET","created_at":"2026-05-28T01:04:16.193533+00:00"},{"alias_kind":"pith_short_16","alias_value":"FO52FBCUDUET6OFT","created_at":"2026-05-28T01:04:16.193533+00:00"},{"alias_kind":"pith_short_8","alias_value":"FO52FBCU","created_at":"2026-05-28T01:04:16.193533+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/FO52FBCUDUET6OFTZ23JRUJA7G","json":"https://pith.science/pith/FO52FBCUDUET6OFTZ23JRUJA7G.json","graph_json":"https://pith.science/api/pith-number/FO52FBCUDUET6OFTZ23JRUJA7G/graph.json","events_json":"https://pith.science/api/pith-number/FO52FBCUDUET6OFTZ23JRUJA7G/events.json","paper":"https://pith.science/paper/FO52FBCU"},"agent_actions":{"view_html":"https://pith.science/pith/FO52FBCUDUET6OFTZ23JRUJA7G","download_json":"https://pith.science/pith/FO52FBCUDUET6OFTZ23JRUJA7G.json","view_paper":"https://pith.science/paper/FO52FBCU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.27569&json=true","fetch_graph":"https://pith.science/api/pith-number/FO52FBCUDUET6OFTZ23JRUJA7G/graph.json","fetch_events":"https://pith.science/api/pith-number/FO52FBCUDUET6OFTZ23JRUJA7G/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FO52FBCUDUET6OFTZ23JRUJA7G/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FO52FBCUDUET6OFTZ23JRUJA7G/action/storage_attestation","attest_author":"https://pith.science/pith/FO52FBCUDUET6OFTZ23JRUJA7G/action/author_attestation","sign_citation":"https://pith.science/pith/FO52FBCUDUET6OFTZ23JRUJA7G/action/citation_signature","submit_replication":"https://pith.science/pith/FO52FBCUDUET6OFTZ23JRUJA7G/action/replication_record"}},"created_at":"2026-05-28T01:04:16.193533+00:00","updated_at":"2026-05-28T01:04:16.193533+00:00"}