{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:SAGV6OD3ZSIKEFNNAAUHBGUM2V","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"cfa91e14d8d62579124724d07933c6fa74127acb045db214d9c30339ca03c446","cross_cats_sorted":["cs.AI","cs.CR"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-26T16:35:59Z","title_canon_sha256":"0c560835a0e0634d4b4b331369fd483a8788e9ab13ca531ea79f1d73f8af97f1"},"schema_version":"1.0","source":{"id":"2402.16717","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.16717","created_at":"2026-07-05T07:49:25Z"},{"alias_kind":"arxiv_version","alias_value":"2402.16717v1","created_at":"2026-07-05T07:49:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.16717","created_at":"2026-07-05T07:49:25Z"},{"alias_kind":"pith_short_12","alias_value":"SAGV6OD3ZSIK","created_at":"2026-07-05T07:49:25Z"},{"alias_kind":"pith_short_16","alias_value":"SAGV6OD3ZSIKEFNN","created_at":"2026-07-05T07:49:25Z"},{"alias_kind":"pith_short_8","alias_value":"SAGV6OD3","created_at":"2026-07-05T07:49:25Z"}],"graph_snapshots":[{"event_id":"sha256:e3d0fdb1513140936426e5238036c9ae369e3b10692be88894a737d78f3a4ac5","target":"graph","created_at":"2026-07-05T07:49:25Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2402.16717/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Adversarial misuse, particularly through `jailbreaking' that circumvents a model's safety and ethical protocols, poses a significant challenge for Large Language Models (LLMs). This paper delves into the mechanisms behind such successful attacks, introducing a hypothesis for the safety mechanism of aligned LLMs: intent security recognition followed by response generation. Grounded in this hypothesis, we propose CodeChameleon, a novel jailbreak framework based on personalized encryption tactics. To elude the intent security recognition phase, we reformulate tasks into a code completion format, ","authors_text":"Caishuang Huang, Huijie Lv, Junjie Ye, Qi Zhang, Shihan Dou, Tao Gui, Xiao Wang, Xuanjing Huang, Yuansen Zhang","cross_cats":["cs.AI","cs.CR"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-26T16:35:59Z","title":"CodeChameleon: Personalized Encryption Framework for Jailbreaking Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.16717","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:9160d058bcb70850d33ccde70477daa7c77899b2fca32d1569a11202f42e4784","target":"record","created_at":"2026-07-05T07:49:25Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"cfa91e14d8d62579124724d07933c6fa74127acb045db214d9c30339ca03c446","cross_cats_sorted":["cs.AI","cs.CR"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-26T16:35:59Z","title_canon_sha256":"0c560835a0e0634d4b4b331369fd483a8788e9ab13ca531ea79f1d73f8af97f1"},"schema_version":"1.0","source":{"id":"2402.16717","kind":"arxiv","version":1}},"canonical_sha256":"900d5f387bcc90a215ad0028709a8cd54cb42cfca85198cc8fff44e59f1c37ca","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"900d5f387bcc90a215ad0028709a8cd54cb42cfca85198cc8fff44e59f1c37ca","first_computed_at":"2026-07-05T07:49:25.420069Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:49:25.420069Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ZrNt9m6UNVIMFDJUC7MCexogjapwIbnqRQJ0F54CTBqeQuvDAEx1L2dsbJgm25bUNAZv71uvRrZ+uogzUGQmBg==","signature_status":"signed_v1","signed_at":"2026-07-05T07:49:25.420501Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.16717","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9160d058bcb70850d33ccde70477daa7c77899b2fca32d1569a11202f42e4784","sha256:e3d0fdb1513140936426e5238036c9ae369e3b10692be88894a737d78f3a4ac5"],"state_sha256":"f8c3e87d07c61579045c08538929371db50c18c90289e66e0771db582ffc8bea"}