{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:PMQ63XPOMJ7VMZANH6CIKTLZTL","short_pith_number":"pith:PMQ63XPO","schema_version":"1.0","canonical_sha256":"7b21edddee627f56640d3f84854d799aca9604880837abbc1de076f604098be3","source":{"kind":"arxiv","id":"2409.11844","version":1},"attestation_state":"computed","paper":{"title":"MEOW: MEMOry Supervised LLM Unlearning Via Inverted Facts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Kexin Huang, Ruilin Luo, Tianle Gu, Yan Teng, Yingchun Wang, Yuanqi Yao, Yujiu Yang","submitted_at":"2024-09-18T09:55:48Z","abstract_excerpt":"Large Language Models (LLMs) can memorize sensitive information, raising concerns about potential misuse. LLM Unlearning, a post-hoc approach to remove this information from trained LLMs, offers a promising solution to mitigate these risks. However, previous practices face three key challenges: 1. Utility: successful unlearning often causes catastrophic collapse on unrelated tasks. 2. Efficiency: many methods either involve adding similarly sized models, which slows down unlearning or inference, or require retain data that are difficult to obtain. 3. Robustness: even effective methods may stil"},"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":"2409.11844","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-09-18T09:55:48Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ba0718101b2043000b53a741ab21b0230865379d7bd1c39ddaa6ae3dac890148","abstract_canon_sha256":"93682e91df73cdd7c72ba0841e8dd44308566b1de7d65f61f85f4b5716dea7ea"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:08:38.744996Z","signature_b64":"bLuDBK9qpFoxe2zo1dcXKeN6iZisvNxCkdedWn1WJdkkdqLK5QlIHkg8SyODblf5bbxcaSnKTmaH/LE+DRzSBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7b21edddee627f56640d3f84854d799aca9604880837abbc1de076f604098be3","last_reissued_at":"2026-07-05T09:08:38.744524Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:08:38.744524Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MEOW: MEMOry Supervised LLM Unlearning Via Inverted Facts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Kexin Huang, Ruilin Luo, Tianle Gu, Yan Teng, Yingchun Wang, Yuanqi Yao, Yujiu Yang","submitted_at":"2024-09-18T09:55:48Z","abstract_excerpt":"Large Language Models (LLMs) can memorize sensitive information, raising concerns about potential misuse. LLM Unlearning, a post-hoc approach to remove this information from trained LLMs, offers a promising solution to mitigate these risks. However, previous practices face three key challenges: 1. Utility: successful unlearning often causes catastrophic collapse on unrelated tasks. 2. Efficiency: many methods either involve adding similarly sized models, which slows down unlearning or inference, or require retain data that are difficult to obtain. 3. Robustness: even effective methods may stil"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.11844","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/2409.11844/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":"2409.11844","created_at":"2026-07-05T09:08:38.744590+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.11844v1","created_at":"2026-07-05T09:08:38.744590+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.11844","created_at":"2026-07-05T09:08:38.744590+00:00"},{"alias_kind":"pith_short_12","alias_value":"PMQ63XPOMJ7V","created_at":"2026-07-05T09:08:38.744590+00:00"},{"alias_kind":"pith_short_16","alias_value":"PMQ63XPOMJ7VMZAN","created_at":"2026-07-05T09:08:38.744590+00:00"},{"alias_kind":"pith_short_8","alias_value":"PMQ63XPO","created_at":"2026-07-05T09:08:38.744590+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.27379","citing_title":"Position: The Term \"Machine Unlearning\" Is Overused in LLMs","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2510.17210","citing_title":"Wisdom is Knowing What not to Say: Hallucination-Free LLMs Unlearning via Attention Shifting","ref_index":27,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PMQ63XPOMJ7VMZANH6CIKTLZTL","json":"https://pith.science/pith/PMQ63XPOMJ7VMZANH6CIKTLZTL.json","graph_json":"https://pith.science/api/pith-number/PMQ63XPOMJ7VMZANH6CIKTLZTL/graph.json","events_json":"https://pith.science/api/pith-number/PMQ63XPOMJ7VMZANH6CIKTLZTL/events.json","paper":"https://pith.science/paper/PMQ63XPO"},"agent_actions":{"view_html":"https://pith.science/pith/PMQ63XPOMJ7VMZANH6CIKTLZTL","download_json":"https://pith.science/pith/PMQ63XPOMJ7VMZANH6CIKTLZTL.json","view_paper":"https://pith.science/paper/PMQ63XPO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.11844&json=true","fetch_graph":"https://pith.science/api/pith-number/PMQ63XPOMJ7VMZANH6CIKTLZTL/graph.json","fetch_events":"https://pith.science/api/pith-number/PMQ63XPOMJ7VMZANH6CIKTLZTL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PMQ63XPOMJ7VMZANH6CIKTLZTL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PMQ63XPOMJ7VMZANH6CIKTLZTL/action/storage_attestation","attest_author":"https://pith.science/pith/PMQ63XPOMJ7VMZANH6CIKTLZTL/action/author_attestation","sign_citation":"https://pith.science/pith/PMQ63XPOMJ7VMZANH6CIKTLZTL/action/citation_signature","submit_replication":"https://pith.science/pith/PMQ63XPOMJ7VMZANH6CIKTLZTL/action/replication_record"}},"created_at":"2026-07-05T09:08:38.744590+00:00","updated_at":"2026-07-05T09:08:38.744590+00:00"}