{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:SK2OTKPI7HTC45UNATYZ3U5W6Z","short_pith_number":"pith:SK2OTKPI","schema_version":"1.0","canonical_sha256":"92b4e9a9e8f9e62e768d04f19dd3b6f6526934ea6d128a69a83f15ebb2d44c4a","source":{"kind":"arxiv","id":"2403.13355","version":1},"attestation_state":"computed","paper":{"title":"BadEdit: Backdooring large language models by model editing","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Jian Zhang, Kangjie Chen, Shangqing Liu, Tianlin Li, Tianwei Zhang, Wenhan Wang, Yang Liu, Yanzhou Li","submitted_at":"2024-03-20T07:34:18Z","abstract_excerpt":"Mainstream backdoor attack methods typically demand substantial tuning data for poisoning, limiting their practicality and potentially degrading the overall performance when applied to Large Language Models (LLMs). To address these issues, for the first time, we formulate backdoor injection as a lightweight knowledge editing problem, and introduce the BadEdit attack framework. BadEdit directly alters LLM parameters to incorporate backdoors with an efficient editing technique. It boasts superiority over existing backdoor injection techniques in several areas: (1) Practicality: BadEdit necessita"},"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":"2403.13355","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-03-20T07:34:18Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"08f7aa7140fa5ef9711424780e176dd7bcc2a8b0678bf46c1b72f8ece0bb8b3e","abstract_canon_sha256":"d37b699deef63063b0ff3f57bb2f696473a1de0e79bed53bb4fed7f245a5419e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:58:35.274592Z","signature_b64":"E1DIB7ywpjHuywAVCMsIDyhcwnzN3xSPUis2h6SFdQgH5q1T+cJ1BU1W1pIJ6LM3rduS9eFfxuCj5/JTK+FpCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"92b4e9a9e8f9e62e768d04f19dd3b6f6526934ea6d128a69a83f15ebb2d44c4a","last_reissued_at":"2026-07-05T07:58:35.274092Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:58:35.274092Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"BadEdit: Backdooring large language models by model editing","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Jian Zhang, Kangjie Chen, Shangqing Liu, Tianlin Li, Tianwei Zhang, Wenhan Wang, Yang Liu, Yanzhou Li","submitted_at":"2024-03-20T07:34:18Z","abstract_excerpt":"Mainstream backdoor attack methods typically demand substantial tuning data for poisoning, limiting their practicality and potentially degrading the overall performance when applied to Large Language Models (LLMs). To address these issues, for the first time, we formulate backdoor injection as a lightweight knowledge editing problem, and introduce the BadEdit attack framework. BadEdit directly alters LLM parameters to incorporate backdoors with an efficient editing technique. It boasts superiority over existing backdoor injection techniques in several areas: (1) Practicality: BadEdit necessita"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.13355","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/2403.13355/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":"2403.13355","created_at":"2026-07-05T07:58:35.274154+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.13355v1","created_at":"2026-07-05T07:58:35.274154+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.13355","created_at":"2026-07-05T07:58:35.274154+00:00"},{"alias_kind":"pith_short_12","alias_value":"SK2OTKPI7HTC","created_at":"2026-07-05T07:58:35.274154+00:00"},{"alias_kind":"pith_short_16","alias_value":"SK2OTKPI7HTC45UN","created_at":"2026-07-05T07:58:35.274154+00:00"},{"alias_kind":"pith_short_8","alias_value":"SK2OTKPI","created_at":"2026-07-05T07:58:35.274154+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.07963","citing_title":"Shared Latent Structures Enable Unified Backdoor Detection and Mitigation in LLMs","ref_index":67,"is_internal_anchor":false},{"citing_arxiv_id":"2606.27511","citing_title":"When the Aggregator Cheats: Data-Free Backdoors in Federated LLM-based QA Systems","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2409.18169","citing_title":"Harmful Fine-tuning Attacks and Defenses for Large Language Models: A Survey","ref_index":93,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12529","citing_title":"BackFlush: Knowledge-Free Backdoor Detection and Elimination with Watermark Preservation in Large Language Models","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2410.02644","citing_title":"Agent Security Bench (ASB): Formalizing and Benchmarking Attacks and Defenses in LLM-based Agents","ref_index":118,"is_internal_anchor":false},{"citing_arxiv_id":"2604.24162","citing_title":"Defusing the Trigger: Plug-and-Play Defense for Backdoored LLMs via Tail-Risk Intrinsic Geometric Smoothing","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2604.21700","citing_title":"Stealthy Backdoor Attacks against LLMs Based on Natural Style Triggers","ref_index":28,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SK2OTKPI7HTC45UNATYZ3U5W6Z","json":"https://pith.science/pith/SK2OTKPI7HTC45UNATYZ3U5W6Z.json","graph_json":"https://pith.science/api/pith-number/SK2OTKPI7HTC45UNATYZ3U5W6Z/graph.json","events_json":"https://pith.science/api/pith-number/SK2OTKPI7HTC45UNATYZ3U5W6Z/events.json","paper":"https://pith.science/paper/SK2OTKPI"},"agent_actions":{"view_html":"https://pith.science/pith/SK2OTKPI7HTC45UNATYZ3U5W6Z","download_json":"https://pith.science/pith/SK2OTKPI7HTC45UNATYZ3U5W6Z.json","view_paper":"https://pith.science/paper/SK2OTKPI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.13355&json=true","fetch_graph":"https://pith.science/api/pith-number/SK2OTKPI7HTC45UNATYZ3U5W6Z/graph.json","fetch_events":"https://pith.science/api/pith-number/SK2OTKPI7HTC45UNATYZ3U5W6Z/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SK2OTKPI7HTC45UNATYZ3U5W6Z/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SK2OTKPI7HTC45UNATYZ3U5W6Z/action/storage_attestation","attest_author":"https://pith.science/pith/SK2OTKPI7HTC45UNATYZ3U5W6Z/action/author_attestation","sign_citation":"https://pith.science/pith/SK2OTKPI7HTC45UNATYZ3U5W6Z/action/citation_signature","submit_replication":"https://pith.science/pith/SK2OTKPI7HTC45UNATYZ3U5W6Z/action/replication_record"}},"created_at":"2026-07-05T07:58:35.274154+00:00","updated_at":"2026-07-05T07:58:35.274154+00:00"}