{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:7UM7TML576TVVWVCGLSJXABKI3","short_pith_number":"pith:7UM7TML5","schema_version":"1.0","canonical_sha256":"fd19f9b17dffa75adaa232e49b802a46d48aebe1870f4741e5c4f721bb22b94b","source":{"kind":"arxiv","id":"2507.21738","version":1},"attestation_state":"computed","paper":{"title":"Zero-Shot Machine Unlearning with Proxy Adversarial Data Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Huiqiang Chen, Tianqing Zhu, Wanlei Zhou, Xin Yu","submitted_at":"2025-07-29T12:16:55Z","abstract_excerpt":"Machine unlearning aims to remove the influence of specific samples from a trained model. A key challenge in this process is over-unlearning, where the model's performance on the remaining data significantly drops due to the change in the model's parameters. Existing unlearning algorithms depend on the remaining data to prevent this issue. As such, these methods are inapplicable in a more practical scenario, where only the unlearning samples are available (i.e., zero-shot unlearning). This paper presents a novel framework, ZS-PAG, to fill this gap. Our approach offers three key innovations: (1"},"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":"2507.21738","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-29T12:16:55Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ac99d0edf5538b35cb371db7f9217ee8fae9575caa1d39cb1a3ad173550b1a69","abstract_canon_sha256":"42f5df535ab7786e77ae07d4eef26ecee830df4f423f2e9b8f32ae55b45f8efa"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:45:05.569258Z","signature_b64":"+zXg5e0sTTNLwi/sqkeV3DIF3fFEFJS+KcYxDtFfsuqZG7Q4x/+H2rnKJB2b73GHl4njBxHIZV3qr5O6mejgAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fd19f9b17dffa75adaa232e49b802a46d48aebe1870f4741e5c4f721bb22b94b","last_reissued_at":"2026-07-05T11:45:05.568881Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:45:05.568881Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Zero-Shot Machine Unlearning with Proxy Adversarial Data Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Huiqiang Chen, Tianqing Zhu, Wanlei Zhou, Xin Yu","submitted_at":"2025-07-29T12:16:55Z","abstract_excerpt":"Machine unlearning aims to remove the influence of specific samples from a trained model. A key challenge in this process is over-unlearning, where the model's performance on the remaining data significantly drops due to the change in the model's parameters. Existing unlearning algorithms depend on the remaining data to prevent this issue. As such, these methods are inapplicable in a more practical scenario, where only the unlearning samples are available (i.e., zero-shot unlearning). This paper presents a novel framework, ZS-PAG, to fill this gap. Our approach offers three key innovations: (1"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.21738","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/2507.21738/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":"2507.21738","created_at":"2026-07-05T11:45:05.568940+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.21738v1","created_at":"2026-07-05T11:45:05.568940+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.21738","created_at":"2026-07-05T11:45:05.568940+00:00"},{"alias_kind":"pith_short_12","alias_value":"7UM7TML576TV","created_at":"2026-07-05T11:45:05.568940+00:00"},{"alias_kind":"pith_short_16","alias_value":"7UM7TML576TVVWVC","created_at":"2026-07-05T11:45:05.568940+00:00"},{"alias_kind":"pith_short_8","alias_value":"7UM7TML5","created_at":"2026-07-05T11:45:05.568940+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/7UM7TML576TVVWVCGLSJXABKI3","json":"https://pith.science/pith/7UM7TML576TVVWVCGLSJXABKI3.json","graph_json":"https://pith.science/api/pith-number/7UM7TML576TVVWVCGLSJXABKI3/graph.json","events_json":"https://pith.science/api/pith-number/7UM7TML576TVVWVCGLSJXABKI3/events.json","paper":"https://pith.science/paper/7UM7TML5"},"agent_actions":{"view_html":"https://pith.science/pith/7UM7TML576TVVWVCGLSJXABKI3","download_json":"https://pith.science/pith/7UM7TML576TVVWVCGLSJXABKI3.json","view_paper":"https://pith.science/paper/7UM7TML5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.21738&json=true","fetch_graph":"https://pith.science/api/pith-number/7UM7TML576TVVWVCGLSJXABKI3/graph.json","fetch_events":"https://pith.science/api/pith-number/7UM7TML576TVVWVCGLSJXABKI3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7UM7TML576TVVWVCGLSJXABKI3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7UM7TML576TVVWVCGLSJXABKI3/action/storage_attestation","attest_author":"https://pith.science/pith/7UM7TML576TVVWVCGLSJXABKI3/action/author_attestation","sign_citation":"https://pith.science/pith/7UM7TML576TVVWVCGLSJXABKI3/action/citation_signature","submit_replication":"https://pith.science/pith/7UM7TML576TVVWVCGLSJXABKI3/action/replication_record"}},"created_at":"2026-07-05T11:45:05.568940+00:00","updated_at":"2026-07-05T11:45:05.568940+00:00"}