{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:KECXRR77ZDBDHLHSAR6HMNCO7S","short_pith_number":"pith:KECXRR77","schema_version":"1.0","canonical_sha256":"510578c7ffc8c233acf2047c76344efca2342f8c44e3a45acb46069289f88b43","source":{"kind":"arxiv","id":"2412.16780","version":4},"attestation_state":"computed","paper":{"title":"Forget Vectors at Play: Universal Input Perturbations Driving Machine Unlearning in Image Classification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Changchang Sun, Gaowen Liu, Jiancheng Liu, Jinghan Jia, Ren Wang, Sijia Liu, Yan Yan, Yihua Zhang","submitted_at":"2024-12-21T21:27:22Z","abstract_excerpt":"Machine unlearning (MU), which seeks to erase the influence of specific unwanted data from already-trained models, is becoming increasingly vital in model editing, particularly to comply with evolving data regulations like the ``right to be forgotten''. Conventional approaches are predominantly model-based, typically requiring retraining or fine-tuning the model's weights to meet unlearning requirements. In this work, we approach the MU problem from a novel input perturbation-based perspective, where the model weights remain intact throughout the unlearning process. We demonstrate the existenc"},"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":"2412.16780","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-12-21T21:27:22Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"4d2e2fa6699bad691291871157bd62d3ebb703377c2bfca9484ad9e4ef19c58f","abstract_canon_sha256":"4f567c16904572ad88d7c5e0e62d6d2d3899b486778635d234b5dddeea738665"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:32:35.973119Z","signature_b64":"tZt8lQoq3a9FgxRWvwxEJJN+L1ZoXHc5EutbZUvmX0xlUZwMto3qFnKadijH5ZiFC5zoFBj7feL8BWLvXPwbDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"510578c7ffc8c233acf2047c76344efca2342f8c44e3a45acb46069289f88b43","last_reissued_at":"2026-07-05T10:32:35.972144Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:32:35.972144Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Forget Vectors at Play: Universal Input Perturbations Driving Machine Unlearning in Image Classification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Changchang Sun, Gaowen Liu, Jiancheng Liu, Jinghan Jia, Ren Wang, Sijia Liu, Yan Yan, Yihua Zhang","submitted_at":"2024-12-21T21:27:22Z","abstract_excerpt":"Machine unlearning (MU), which seeks to erase the influence of specific unwanted data from already-trained models, is becoming increasingly vital in model editing, particularly to comply with evolving data regulations like the ``right to be forgotten''. Conventional approaches are predominantly model-based, typically requiring retraining or fine-tuning the model's weights to meet unlearning requirements. In this work, we approach the MU problem from a novel input perturbation-based perspective, where the model weights remain intact throughout the unlearning process. We demonstrate the existenc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.16780","kind":"arxiv","version":4},"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/2412.16780/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":"2412.16780","created_at":"2026-07-05T10:32:35.972285+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.16780v4","created_at":"2026-07-05T10:32:35.972285+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.16780","created_at":"2026-07-05T10:32:35.972285+00:00"},{"alias_kind":"pith_short_12","alias_value":"KECXRR77ZDBD","created_at":"2026-07-05T10:32:35.972285+00:00"},{"alias_kind":"pith_short_16","alias_value":"KECXRR77ZDBDHLHS","created_at":"2026-07-05T10:32:35.972285+00:00"},{"alias_kind":"pith_short_8","alias_value":"KECXRR77","created_at":"2026-07-05T10:32:35.972285+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.24022","citing_title":"IPRU: Input-Perturbation-based Radio Frequency Fingerprinting Unlearning for LAWNs","ref_index":10,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KECXRR77ZDBDHLHSAR6HMNCO7S","json":"https://pith.science/pith/KECXRR77ZDBDHLHSAR6HMNCO7S.json","graph_json":"https://pith.science/api/pith-number/KECXRR77ZDBDHLHSAR6HMNCO7S/graph.json","events_json":"https://pith.science/api/pith-number/KECXRR77ZDBDHLHSAR6HMNCO7S/events.json","paper":"https://pith.science/paper/KECXRR77"},"agent_actions":{"view_html":"https://pith.science/pith/KECXRR77ZDBDHLHSAR6HMNCO7S","download_json":"https://pith.science/pith/KECXRR77ZDBDHLHSAR6HMNCO7S.json","view_paper":"https://pith.science/paper/KECXRR77","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.16780&json=true","fetch_graph":"https://pith.science/api/pith-number/KECXRR77ZDBDHLHSAR6HMNCO7S/graph.json","fetch_events":"https://pith.science/api/pith-number/KECXRR77ZDBDHLHSAR6HMNCO7S/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KECXRR77ZDBDHLHSAR6HMNCO7S/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KECXRR77ZDBDHLHSAR6HMNCO7S/action/storage_attestation","attest_author":"https://pith.science/pith/KECXRR77ZDBDHLHSAR6HMNCO7S/action/author_attestation","sign_citation":"https://pith.science/pith/KECXRR77ZDBDHLHSAR6HMNCO7S/action/citation_signature","submit_replication":"https://pith.science/pith/KECXRR77ZDBDHLHSAR6HMNCO7S/action/replication_record"}},"created_at":"2026-07-05T10:32:35.972285+00:00","updated_at":"2026-07-05T10:32:35.972285+00:00"}