{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:BO2VYWUJMPVI7A36R2DQ7OSQS4","short_pith_number":"pith:BO2VYWUJ","schema_version":"1.0","canonical_sha256":"0bb55c5a8963ea8f837e8e870fba50973c90a09b52be6d91a6befd588943b267","source":{"kind":"arxiv","id":"2406.06822","version":1},"attestation_state":"computed","paper":{"title":"An LLM-Assisted Easy-to-Trigger Backdoor Attack on Code Completion Models: Injecting Disguised Vulnerabilities against Strong Detection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.SE"],"primary_cat":"cs.CR","authors_text":"Doowon Kim, Hanbin Hong, Kiho Lee, Shenao Yan, Shen Wang, Yuan Hong, Yue Duan","submitted_at":"2024-06-10T22:10:05Z","abstract_excerpt":"Large Language Models (LLMs) have transformed code completion tasks, providing context-based suggestions to boost developer productivity in software engineering. As users often fine-tune these models for specific applications, poisoning and backdoor attacks can covertly alter the model outputs. To address this critical security challenge, we introduce CodeBreaker, a pioneering LLM-assisted backdoor attack framework on code completion models. Unlike recent attacks that embed malicious payloads in detectable or irrelevant sections of the code (e.g., comments), CodeBreaker leverages LLMs (e.g., G"},"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":"2406.06822","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2024-06-10T22:10:05Z","cross_cats_sorted":["cs.AI","cs.SE"],"title_canon_sha256":"848b48e3a395b6701e441516db3d285630308630af80ddbca094c3599ec7eb3d","abstract_canon_sha256":"a3306c653b4f1cbe83c7406ed4a09e44c3812a91ebf5ace84475b3832c22846a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:29:50.532257Z","signature_b64":"lJA2dL5PdHF3c2eSjVW80KxcuLNemFIJWtEMRuxnFYBIQTgxYMJyAlPlt7nvsoz7QuwJ5jOfyUjVXn/yS+gxDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0bb55c5a8963ea8f837e8e870fba50973c90a09b52be6d91a6befd588943b267","last_reissued_at":"2026-07-05T08:29:50.531844Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:29:50.531844Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"An LLM-Assisted Easy-to-Trigger Backdoor Attack on Code Completion Models: Injecting Disguised Vulnerabilities against Strong Detection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.SE"],"primary_cat":"cs.CR","authors_text":"Doowon Kim, Hanbin Hong, Kiho Lee, Shenao Yan, Shen Wang, Yuan Hong, Yue Duan","submitted_at":"2024-06-10T22:10:05Z","abstract_excerpt":"Large Language Models (LLMs) have transformed code completion tasks, providing context-based suggestions to boost developer productivity in software engineering. As users often fine-tune these models for specific applications, poisoning and backdoor attacks can covertly alter the model outputs. To address this critical security challenge, we introduce CodeBreaker, a pioneering LLM-assisted backdoor attack framework on code completion models. Unlike recent attacks that embed malicious payloads in detectable or irrelevant sections of the code (e.g., comments), CodeBreaker leverages LLMs (e.g., G"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.06822","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/2406.06822/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":"2406.06822","created_at":"2026-07-05T08:29:50.531900+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.06822v1","created_at":"2026-07-05T08:29:50.531900+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.06822","created_at":"2026-07-05T08:29:50.531900+00:00"},{"alias_kind":"pith_short_12","alias_value":"BO2VYWUJMPVI","created_at":"2026-07-05T08:29:50.531900+00:00"},{"alias_kind":"pith_short_16","alias_value":"BO2VYWUJMPVI7A36","created_at":"2026-07-05T08:29:50.531900+00:00"},{"alias_kind":"pith_short_8","alias_value":"BO2VYWUJ","created_at":"2026-07-05T08:29:50.531900+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2510.04265","citing_title":"Don't Pass@k: A Bayesian Framework for Large Language Model Evaluation","ref_index":63,"is_internal_anchor":false},{"citing_arxiv_id":"2510.16558","citing_title":"A First Look at the Security Issues in the Model Context Protocol Ecosystem","ref_index":53,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BO2VYWUJMPVI7A36R2DQ7OSQS4","json":"https://pith.science/pith/BO2VYWUJMPVI7A36R2DQ7OSQS4.json","graph_json":"https://pith.science/api/pith-number/BO2VYWUJMPVI7A36R2DQ7OSQS4/graph.json","events_json":"https://pith.science/api/pith-number/BO2VYWUJMPVI7A36R2DQ7OSQS4/events.json","paper":"https://pith.science/paper/BO2VYWUJ"},"agent_actions":{"view_html":"https://pith.science/pith/BO2VYWUJMPVI7A36R2DQ7OSQS4","download_json":"https://pith.science/pith/BO2VYWUJMPVI7A36R2DQ7OSQS4.json","view_paper":"https://pith.science/paper/BO2VYWUJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.06822&json=true","fetch_graph":"https://pith.science/api/pith-number/BO2VYWUJMPVI7A36R2DQ7OSQS4/graph.json","fetch_events":"https://pith.science/api/pith-number/BO2VYWUJMPVI7A36R2DQ7OSQS4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BO2VYWUJMPVI7A36R2DQ7OSQS4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BO2VYWUJMPVI7A36R2DQ7OSQS4/action/storage_attestation","attest_author":"https://pith.science/pith/BO2VYWUJMPVI7A36R2DQ7OSQS4/action/author_attestation","sign_citation":"https://pith.science/pith/BO2VYWUJMPVI7A36R2DQ7OSQS4/action/citation_signature","submit_replication":"https://pith.science/pith/BO2VYWUJMPVI7A36R2DQ7OSQS4/action/replication_record"}},"created_at":"2026-07-05T08:29:50.531900+00:00","updated_at":"2026-07-05T08:29:50.531900+00:00"}