{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:BVGCRLCXZNMBBUTDPOYVXXOJYU","short_pith_number":"pith:BVGCRLCX","schema_version":"1.0","canonical_sha256":"0d4c28ac57cb5810d2637bb15bddc9c51fe9399f07858c6115b7acf129017c8c","source":{"kind":"arxiv","id":"2607.23934","version":1},"attestation_state":"computed","paper":{"title":"DECAF: De-Clustering for Adaptive Representational Unlearning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Anjie Le, Can Peng, Hongcheng Guo, J. Alison Noble","submitted_at":"2026-07-27T02:11:58Z","abstract_excerpt":"Machine unlearning, which aims to remove the influence of specific training data from a trained model, is a key requirement for privacy, accountability, and adaptive deployment. We argue that many unlearning methods are vulnerable to a simple clustering attack, which can recover class structure in an unsupervised manner, limiting their suitability for continual deployment where removal requests must be handled reliably on demand. To address this, we propose DECAF (DE-Clustering for Adaptive Forgetting), a post-hoc method that operates only on the forget set and is designed to break the cluster"},"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":"2607.23934","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-27T02:11:58Z","cross_cats_sorted":[],"title_canon_sha256":"32e9a763b390b0e4ad1eae956fb7b57e2a3dfc7431460f1a17dfb81cc3d3fa4b","abstract_canon_sha256":"8bae8800aee50bc7d71056c8967bcf5ae6dce1752d2b7eaaf4b1bf1efab4e991"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-28T01:23:36.489380Z","signature_b64":"WBKF4NjX3WJwEJFGz7rr9Rs0rPuD66Fx5+sC5wFU9Zb7gr1XSv/AUIn4QHZe2Rwr927kmUlrdePMK5meLAicBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0d4c28ac57cb5810d2637bb15bddc9c51fe9399f07858c6115b7acf129017c8c","last_reissued_at":"2026-07-28T01:23:36.488539Z","signature_status":"signed_v1","first_computed_at":"2026-07-28T01:23:36.488539Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DECAF: De-Clustering for Adaptive Representational Unlearning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Anjie Le, Can Peng, Hongcheng Guo, J. Alison Noble","submitted_at":"2026-07-27T02:11:58Z","abstract_excerpt":"Machine unlearning, which aims to remove the influence of specific training data from a trained model, is a key requirement for privacy, accountability, and adaptive deployment. We argue that many unlearning methods are vulnerable to a simple clustering attack, which can recover class structure in an unsupervised manner, limiting their suitability for continual deployment where removal requests must be handled reliably on demand. To address this, we propose DECAF (DE-Clustering for Adaptive Forgetting), a post-hoc method that operates only on the forget set and is designed to break the cluster"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.23934","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/2607.23934/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":"2607.23934","created_at":"2026-07-28T01:23:36.488959+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.23934v1","created_at":"2026-07-28T01:23:36.488959+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.23934","created_at":"2026-07-28T01:23:36.488959+00:00"},{"alias_kind":"pith_short_12","alias_value":"BVGCRLCXZNMB","created_at":"2026-07-28T01:23:36.488959+00:00"},{"alias_kind":"pith_short_16","alias_value":"BVGCRLCXZNMBBUTD","created_at":"2026-07-28T01:23:36.488959+00:00"},{"alias_kind":"pith_short_8","alias_value":"BVGCRLCX","created_at":"2026-07-28T01:23:36.488959+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/BVGCRLCXZNMBBUTDPOYVXXOJYU","json":"https://pith.science/pith/BVGCRLCXZNMBBUTDPOYVXXOJYU.json","graph_json":"https://pith.science/api/pith-number/BVGCRLCXZNMBBUTDPOYVXXOJYU/graph.json","events_json":"https://pith.science/api/pith-number/BVGCRLCXZNMBBUTDPOYVXXOJYU/events.json","paper":"https://pith.science/paper/BVGCRLCX"},"agent_actions":{"view_html":"https://pith.science/pith/BVGCRLCXZNMBBUTDPOYVXXOJYU","download_json":"https://pith.science/pith/BVGCRLCXZNMBBUTDPOYVXXOJYU.json","view_paper":"https://pith.science/paper/BVGCRLCX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.23934&json=true","fetch_graph":"https://pith.science/api/pith-number/BVGCRLCXZNMBBUTDPOYVXXOJYU/graph.json","fetch_events":"https://pith.science/api/pith-number/BVGCRLCXZNMBBUTDPOYVXXOJYU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BVGCRLCXZNMBBUTDPOYVXXOJYU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BVGCRLCXZNMBBUTDPOYVXXOJYU/action/storage_attestation","attest_author":"https://pith.science/pith/BVGCRLCXZNMBBUTDPOYVXXOJYU/action/author_attestation","sign_citation":"https://pith.science/pith/BVGCRLCXZNMBBUTDPOYVXXOJYU/action/citation_signature","submit_replication":"https://pith.science/pith/BVGCRLCXZNMBBUTDPOYVXXOJYU/action/replication_record"}},"created_at":"2026-07-28T01:23:36.488959+00:00","updated_at":"2026-07-28T01:23:36.488959+00:00"}