{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:H3XL25PSZ55URGBK6V45DZLPZT","short_pith_number":"pith:H3XL25PS","canonical_record":{"source":{"id":"2409.14177","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2024-09-21T15:36:26Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2c3e53be1996d412e3d7eceb321f2a60a62868f61e06b45faf66e5fcc4dec165","abstract_canon_sha256":"62cda2b0a1e6ba77be739339568c493f081a5601d1a38134624c11953cdaec5d"},"schema_version":"1.0"},"canonical_sha256":"3eeebd75f2cf7b48982af579d1e56fccf27c31102a7b8883f9d165ef76a4cc95","source":{"kind":"arxiv","id":"2409.14177","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2409.14177","created_at":"2026-07-05T09:15:14Z"},{"alias_kind":"arxiv_version","alias_value":"2409.14177v2","created_at":"2026-07-05T09:15:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.14177","created_at":"2026-07-05T09:15:14Z"},{"alias_kind":"pith_short_12","alias_value":"H3XL25PSZ55U","created_at":"2026-07-05T09:15:14Z"},{"alias_kind":"pith_short_16","alias_value":"H3XL25PSZ55URGBK","created_at":"2026-07-05T09:15:14Z"},{"alias_kind":"pith_short_8","alias_value":"H3XL25PS","created_at":"2026-07-05T09:15:14Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:H3XL25PSZ55URGBK6V45DZLPZT","target":"record","payload":{"canonical_record":{"source":{"id":"2409.14177","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2024-09-21T15:36:26Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2c3e53be1996d412e3d7eceb321f2a60a62868f61e06b45faf66e5fcc4dec165","abstract_canon_sha256":"62cda2b0a1e6ba77be739339568c493f081a5601d1a38134624c11953cdaec5d"},"schema_version":"1.0"},"canonical_sha256":"3eeebd75f2cf7b48982af579d1e56fccf27c31102a7b8883f9d165ef76a4cc95","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:15:14.687627Z","signature_b64":"juJW2tIHG3x6DZ6LIbIfoZFVfT8iBcT3PMbSA1DA8wWC+bdqN8ULzpxlUy+7Vwbp5kA0gBJ5++3dGyZpmSElAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3eeebd75f2cf7b48982af579d1e56fccf27c31102a7b8883f9d165ef76a4cc95","last_reissued_at":"2026-07-05T09:15:14.687129Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:15:14.687129Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2409.14177","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:15:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CEFEjKSds/wGrqLLW9F/MuibmUMCj9pmoQutGdx0KvyN6RuBp1oJ/MeUpiQQtr5lWo5FtjvVtgTyIFGYHZk4Dw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T07:24:58.343805Z"},"content_sha256":"be79f21e9a2d32dee7b157384610df54c441a72539767f017f4888a1bca743e4","schema_version":"1.0","event_id":"sha256:be79f21e9a2d32dee7b157384610df54c441a72539767f017f4888a1bca743e4"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:H3XL25PSZ55URGBK6V45DZLPZT","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"PathSeeker: Exploring LLM Security Vulnerabilities with a Reinforcement Learning-Based Jailbreak Approach","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Haoyu Wang, Jun Wang, Li Li, Mingyi Zhou, Wei Ma, Yang Liu, Yanjie Zhao, Zhihao Lin","submitted_at":"2024-09-21T15:36:26Z","abstract_excerpt":"In recent years, Large Language Models (LLMs) have gained widespread use, raising concerns about their security. Traditional jailbreak attacks, which often rely on the model internal information or have limitations when exploring the unsafe behavior of the victim model, limiting their reducing their general applicability. In this paper, we introduce PathSeeker, a novel black-box jailbreak method, which is inspired by the game of rats escaping a maze. We think that each LLM has its unique \"security maze\", and attackers attempt to find the exit learning from the received feedback and their accum"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.14177","kind":"arxiv","version":2},"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/2409.14177/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":"2026-07-05T09:15:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IJDwohic36RDlz04HUqqy2byW9toOc3J7KGUkQeGOQj2/0ZVu6K4xhmWjIpXfrs1VXm1yc/YEni04VaSZ4mXBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T07:24:58.344391Z"},"content_sha256":"c47435ef139b2d8f68cd9ba4a2e99f31d8bacab194ba532ba1097554b0544c1f","schema_version":"1.0","event_id":"sha256:c47435ef139b2d8f68cd9ba4a2e99f31d8bacab194ba532ba1097554b0544c1f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/H3XL25PSZ55URGBK6V45DZLPZT/bundle.json","state_url":"https://pith.science/pith/H3XL25PSZ55URGBK6V45DZLPZT/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/H3XL25PSZ55URGBK6V45DZLPZT/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":"2026-08-14T07:24:58Z","links":{"resolver":"https://pith.science/pith/H3XL25PSZ55URGBK6V45DZLPZT","bundle":"https://pith.science/pith/H3XL25PSZ55URGBK6V45DZLPZT/bundle.json","state":"https://pith.science/pith/H3XL25PSZ55URGBK6V45DZLPZT/state.json","well_known_bundle":"https://pith.science/.well-known/pith/H3XL25PSZ55URGBK6V45DZLPZT/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:H3XL25PSZ55URGBK6V45DZLPZT","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":"62cda2b0a1e6ba77be739339568c493f081a5601d1a38134624c11953cdaec5d","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2024-09-21T15:36:26Z","title_canon_sha256":"2c3e53be1996d412e3d7eceb321f2a60a62868f61e06b45faf66e5fcc4dec165"},"schema_version":"1.0","source":{"id":"2409.14177","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2409.14177","created_at":"2026-07-05T09:15:14Z"},{"alias_kind":"arxiv_version","alias_value":"2409.14177v2","created_at":"2026-07-05T09:15:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.14177","created_at":"2026-07-05T09:15:14Z"},{"alias_kind":"pith_short_12","alias_value":"H3XL25PSZ55U","created_at":"2026-07-05T09:15:14Z"},{"alias_kind":"pith_short_16","alias_value":"H3XL25PSZ55URGBK","created_at":"2026-07-05T09:15:14Z"},{"alias_kind":"pith_short_8","alias_value":"H3XL25PS","created_at":"2026-07-05T09:15:14Z"}],"graph_snapshots":[{"event_id":"sha256:c47435ef139b2d8f68cd9ba4a2e99f31d8bacab194ba532ba1097554b0544c1f","target":"graph","created_at":"2026-07-05T09:15:14Z","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/2409.14177/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"In recent years, Large Language Models (LLMs) have gained widespread use, raising concerns about their security. Traditional jailbreak attacks, which often rely on the model internal information or have limitations when exploring the unsafe behavior of the victim model, limiting their reducing their general applicability. In this paper, we introduce PathSeeker, a novel black-box jailbreak method, which is inspired by the game of rats escaping a maze. We think that each LLM has its unique \"security maze\", and attackers attempt to find the exit learning from the received feedback and their accum","authors_text":"Haoyu Wang, Jun Wang, Li Li, Mingyi Zhou, Wei Ma, Yang Liu, Yanjie Zhao, Zhihao Lin","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2024-09-21T15:36:26Z","title":"PathSeeker: Exploring LLM Security Vulnerabilities with a Reinforcement Learning-Based Jailbreak Approach"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.14177","kind":"arxiv","version":2},"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:be79f21e9a2d32dee7b157384610df54c441a72539767f017f4888a1bca743e4","target":"record","created_at":"2026-07-05T09:15:14Z","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":"62cda2b0a1e6ba77be739339568c493f081a5601d1a38134624c11953cdaec5d","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2024-09-21T15:36:26Z","title_canon_sha256":"2c3e53be1996d412e3d7eceb321f2a60a62868f61e06b45faf66e5fcc4dec165"},"schema_version":"1.0","source":{"id":"2409.14177","kind":"arxiv","version":2}},"canonical_sha256":"3eeebd75f2cf7b48982af579d1e56fccf27c31102a7b8883f9d165ef76a4cc95","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3eeebd75f2cf7b48982af579d1e56fccf27c31102a7b8883f9d165ef76a4cc95","first_computed_at":"2026-07-05T09:15:14.687129Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:15:14.687129Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"juJW2tIHG3x6DZ6LIbIfoZFVfT8iBcT3PMbSA1DA8wWC+bdqN8ULzpxlUy+7Vwbp5kA0gBJ5++3dGyZpmSElAg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:15:14.687627Z","signed_message":"canonical_sha256_bytes"},"source_id":"2409.14177","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:be79f21e9a2d32dee7b157384610df54c441a72539767f017f4888a1bca743e4","sha256:c47435ef139b2d8f68cd9ba4a2e99f31d8bacab194ba532ba1097554b0544c1f"],"state_sha256":"f7faf65141b01fbf58886f31fa19254d3da09cc6ca9180ebbcc2bc6ec9ea8bc1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IXPtI+gcGTf/iDs30ZlpnPm0nZjNnKVl35uO2xRyZY9CDp3FSs0QWUzsIon0UwSS6lhpYo64oFw7OyoqZV6DAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-14T07:24:58.350070Z","bundle_sha256":"988e3de346fa93f7994041ae9206f839bf10903d579edd87019f4b8094402fde"}}