{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:PCJ34T3SV43VHQDQETRZLGLISU","short_pith_number":"pith:PCJ34T3S","canonical_record":{"source":{"id":"2509.05471","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-09-05T19:57:38Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"617d029458a1c157d03f495c7af3ee47ac5576b4677b11793674f6a9841b3f7d","abstract_canon_sha256":"a28d3e45152895eb6d47233796facbc6f5a05ad9f3d8da84e42391ea42c0ab86"},"schema_version":"1.0"},"canonical_sha256":"7893be4f72af3753c07024e3959968953881a42c48a971255e891e4960bcd639","source":{"kind":"arxiv","id":"2509.05471","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.05471","created_at":"2026-07-05T12:05:48Z"},{"alias_kind":"arxiv_version","alias_value":"2509.05471v1","created_at":"2026-07-05T12:05:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.05471","created_at":"2026-07-05T12:05:48Z"},{"alias_kind":"pith_short_12","alias_value":"PCJ34T3SV43V","created_at":"2026-07-05T12:05:48Z"},{"alias_kind":"pith_short_16","alias_value":"PCJ34T3SV43VHQDQ","created_at":"2026-07-05T12:05:48Z"},{"alias_kind":"pith_short_8","alias_value":"PCJ34T3S","created_at":"2026-07-05T12:05:48Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:PCJ34T3SV43VHQDQETRZLGLISU","target":"record","payload":{"canonical_record":{"source":{"id":"2509.05471","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-09-05T19:57:38Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"617d029458a1c157d03f495c7af3ee47ac5576b4677b11793674f6a9841b3f7d","abstract_canon_sha256":"a28d3e45152895eb6d47233796facbc6f5a05ad9f3d8da84e42391ea42c0ab86"},"schema_version":"1.0"},"canonical_sha256":"7893be4f72af3753c07024e3959968953881a42c48a971255e891e4960bcd639","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:05:48.206391Z","signature_b64":"L+GKuYeJQ6Yu5MmV7WbQWYI2RzMZmH45fJpP3JuaFo8MM5+8V1wUD0P8G/A5Av6DBOpn93I6Mzd8dF0Pa1lnDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7893be4f72af3753c07024e3959968953881a42c48a971255e891e4960bcd639","last_reissued_at":"2026-07-05T12:05:48.205785Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:05:48.205785Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2509.05471","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T12:05:48Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"UqiG+x29GvyFFsRGz0SKhgOyRMA0YaVTxH3B4rcGFkKmQiw0grv+DaxjRdKJpRorBaHTTKV8j5xHzgXqMibPDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T18:42:17.605512Z"},"content_sha256":"e70edb858027a76fe1ec8bf7489fd06d60793a62d8b5685e5b9e5484bc968ebb","schema_version":"1.0","event_id":"sha256:e70edb858027a76fe1ec8bf7489fd06d60793a62d8b5685e5b9e5484bc968ebb"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:PCJ34T3SV43VHQDQETRZLGLISU","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Behind the Mask: Benchmarking Camouflaged Jailbreaks in Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Mohammad Zandsalimy, Shanu Sushmita, Youjia Zheng","submitted_at":"2025-09-05T19:57:38Z","abstract_excerpt":"Large Language Models (LLMs) are increasingly vulnerable to a sophisticated form of adversarial prompting known as camouflaged jailbreaking. This method embeds malicious intent within seemingly benign language to evade existing safety mechanisms. Unlike overt attacks, these subtle prompts exploit contextual ambiguity and the flexible nature of language, posing significant challenges to current defense systems. This paper investigates the construction and impact of camouflaged jailbreak prompts, emphasizing their deceptive characteristics and the limitations of traditional keyword-based detecti"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.05471","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/2509.05471/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T12:05:48Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ANcsVD5Eqn4L4rTuC34PlBZ59xOdgSmdezLLG0f6VulAOPVUPrSU+Wu7Fr7vVtIh6jdgeK2U2BbmKY4S13eTDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T18:42:17.606420Z"},"content_sha256":"944a674645ff2a27377508ab50742aedbf7e676da452a1ce79b48928ac06f626","schema_version":"1.0","event_id":"sha256:944a674645ff2a27377508ab50742aedbf7e676da452a1ce79b48928ac06f626"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/PCJ34T3SV43VHQDQETRZLGLISU/bundle.json","state_url":"https://pith.science/pith/PCJ34T3SV43VHQDQETRZLGLISU/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/PCJ34T3SV43VHQDQETRZLGLISU/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-13T18:42:17Z","links":{"resolver":"https://pith.science/pith/PCJ34T3SV43VHQDQETRZLGLISU","bundle":"https://pith.science/pith/PCJ34T3SV43VHQDQETRZLGLISU/bundle.json","state":"https://pith.science/pith/PCJ34T3SV43VHQDQETRZLGLISU/state.json","well_known_bundle":"https://pith.science/.well-known/pith/PCJ34T3SV43VHQDQETRZLGLISU/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:PCJ34T3SV43VHQDQETRZLGLISU","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":"a28d3e45152895eb6d47233796facbc6f5a05ad9f3d8da84e42391ea42c0ab86","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-09-05T19:57:38Z","title_canon_sha256":"617d029458a1c157d03f495c7af3ee47ac5576b4677b11793674f6a9841b3f7d"},"schema_version":"1.0","source":{"id":"2509.05471","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.05471","created_at":"2026-07-05T12:05:48Z"},{"alias_kind":"arxiv_version","alias_value":"2509.05471v1","created_at":"2026-07-05T12:05:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.05471","created_at":"2026-07-05T12:05:48Z"},{"alias_kind":"pith_short_12","alias_value":"PCJ34T3SV43V","created_at":"2026-07-05T12:05:48Z"},{"alias_kind":"pith_short_16","alias_value":"PCJ34T3SV43VHQDQ","created_at":"2026-07-05T12:05:48Z"},{"alias_kind":"pith_short_8","alias_value":"PCJ34T3S","created_at":"2026-07-05T12:05:48Z"}],"graph_snapshots":[{"event_id":"sha256:944a674645ff2a27377508ab50742aedbf7e676da452a1ce79b48928ac06f626","target":"graph","created_at":"2026-07-05T12:05:48Z","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/2509.05471/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) are increasingly vulnerable to a sophisticated form of adversarial prompting known as camouflaged jailbreaking. This method embeds malicious intent within seemingly benign language to evade existing safety mechanisms. Unlike overt attacks, these subtle prompts exploit contextual ambiguity and the flexible nature of language, posing significant challenges to current defense systems. This paper investigates the construction and impact of camouflaged jailbreak prompts, emphasizing their deceptive characteristics and the limitations of traditional keyword-based detecti","authors_text":"Mohammad Zandsalimy, Shanu Sushmita, Youjia Zheng","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-09-05T19:57:38Z","title":"Behind the Mask: Benchmarking Camouflaged Jailbreaks in Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.05471","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:e70edb858027a76fe1ec8bf7489fd06d60793a62d8b5685e5b9e5484bc968ebb","target":"record","created_at":"2026-07-05T12:05:48Z","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":"a28d3e45152895eb6d47233796facbc6f5a05ad9f3d8da84e42391ea42c0ab86","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-09-05T19:57:38Z","title_canon_sha256":"617d029458a1c157d03f495c7af3ee47ac5576b4677b11793674f6a9841b3f7d"},"schema_version":"1.0","source":{"id":"2509.05471","kind":"arxiv","version":1}},"canonical_sha256":"7893be4f72af3753c07024e3959968953881a42c48a971255e891e4960bcd639","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7893be4f72af3753c07024e3959968953881a42c48a971255e891e4960bcd639","first_computed_at":"2026-07-05T12:05:48.205785Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T12:05:48.205785Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"L+GKuYeJQ6Yu5MmV7WbQWYI2RzMZmH45fJpP3JuaFo8MM5+8V1wUD0P8G/A5Av6DBOpn93I6Mzd8dF0Pa1lnDg==","signature_status":"signed_v1","signed_at":"2026-07-05T12:05:48.206391Z","signed_message":"canonical_sha256_bytes"},"source_id":"2509.05471","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e70edb858027a76fe1ec8bf7489fd06d60793a62d8b5685e5b9e5484bc968ebb","sha256:944a674645ff2a27377508ab50742aedbf7e676da452a1ce79b48928ac06f626"],"state_sha256":"c6cd12025609470f42352afaac1f1c2e6c881d20b93ddac3f51f2a0dcf870438"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"YkwYVbazGc0U/l2krt1Fuspt5+5FjBxTYu9/rI/rs7xGfEjHJR5kJuqkdDQ4H7Vf54/cxjg9mCOQDpvzAkf9Dw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-13T18:42:17.620664Z","bundle_sha256":"1bc2ae401357bbf6e6c3db12fccc947e339226e7fa4b1019dfd2a1a652307041"}}