{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:AMROHAMOZXISYU65HFNNXHOMUY","short_pith_number":"pith:AMROHAMO","canonical_record":{"source":{"id":"2406.08725","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-06-13T01:05:22Z","cross_cats_sorted":[],"title_canon_sha256":"97a1ad1c4f626549e32778ffa4e8068b42cdde735891b461467be4a123b4b695","abstract_canon_sha256":"0dccc61c6d0649a8a2ec94dd0e46c0a7a41fb53916af18f7b42f811f3d874c54"},"schema_version":"1.0"},"canonical_sha256":"0322e3818ecdd12c53dd395adb9dcca631ba002dcb525a5eeccd0c315151a244","source":{"kind":"arxiv","id":"2406.08725","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.08725","created_at":"2026-07-05T08:31:22Z"},{"alias_kind":"arxiv_version","alias_value":"2406.08725v1","created_at":"2026-07-05T08:31:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.08725","created_at":"2026-07-05T08:31:22Z"},{"alias_kind":"pith_short_12","alias_value":"AMROHAMOZXIS","created_at":"2026-07-05T08:31:22Z"},{"alias_kind":"pith_short_16","alias_value":"AMROHAMOZXISYU65","created_at":"2026-07-05T08:31:22Z"},{"alias_kind":"pith_short_8","alias_value":"AMROHAMO","created_at":"2026-07-05T08:31:22Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:AMROHAMOZXISYU65HFNNXHOMUY","target":"record","payload":{"canonical_record":{"source":{"id":"2406.08725","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-06-13T01:05:22Z","cross_cats_sorted":[],"title_canon_sha256":"97a1ad1c4f626549e32778ffa4e8068b42cdde735891b461467be4a123b4b695","abstract_canon_sha256":"0dccc61c6d0649a8a2ec94dd0e46c0a7a41fb53916af18f7b42f811f3d874c54"},"schema_version":"1.0"},"canonical_sha256":"0322e3818ecdd12c53dd395adb9dcca631ba002dcb525a5eeccd0c315151a244","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:31:22.245119Z","signature_b64":"arl9lWDO5XGm/I1w3xyuu++mPtLXVaTe2Sb3nGbX9mtnBvuhFaUyeI2861ugVBrIpDbrZgD6JHUoqlFmo/SOAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0322e3818ecdd12c53dd395adb9dcca631ba002dcb525a5eeccd0c315151a244","last_reissued_at":"2026-07-05T08:31:22.244658Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:31:22.244658Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2406.08725","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":"2026-07-05T08:31:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Z8Ty5WlfeBPWcJxIRUhA8Gxb4wifd6kFvIV6u0Ncrv0Q5rhwUWYF9e3rXmTuRN9dEvyl8bI72QB7M9RB67G9DQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T15:21:27.919392Z"},"content_sha256":"c7683fd69df233227b18ed6f74a40b6429db0c8ddfac2fb175b868ff9c4cbba3","schema_version":"1.0","event_id":"sha256:c7683fd69df233227b18ed6f74a40b6429db0c8ddfac2fb175b868ff9c4cbba3"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:AMROHAMOZXISYU65HFNNXHOMUY","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"RL-JACK: Reinforcement Learning-powered Black-box Jailbreaking Attack against LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CR","authors_text":"Lu Yan, Wenbo Guo, Xiangyu Zhang, Xuan Chen, Yunshu Mao, Yuzhou Nie","submitted_at":"2024-06-13T01:05:22Z","abstract_excerpt":"Modern large language model (LLM) developers typically conduct a safety alignment to prevent an LLM from generating unethical or harmful content. Recent studies have discovered that the safety alignment of LLMs can be bypassed by jailbreaking prompts. These prompts are designed to create specific conversation scenarios with a harmful question embedded. Querying an LLM with such prompts can mislead the model into responding to the harmful question. The stochastic and random nature of existing genetic methods largely limits the effectiveness and efficiency of state-of-the-art (SOTA) jailbreaking"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.08725","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/2406.08725/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":"2026-07-05T08:31:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"N8FsKJj/QhxhL72qXNAg0QRi8Pz9tJL21puS8Jwla0cclFnUmBaSe3FxcKZVWY5KqJy84fkj1N3xNaBWLAnWDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T15:21:27.919877Z"},"content_sha256":"3a032d154e486417a9cbd886934bcb7ca9e76299a5abde247288546219c301d2","schema_version":"1.0","event_id":"sha256:3a032d154e486417a9cbd886934bcb7ca9e76299a5abde247288546219c301d2"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/AMROHAMOZXISYU65HFNNXHOMUY/bundle.json","state_url":"https://pith.science/pith/AMROHAMOZXISYU65HFNNXHOMUY/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/AMROHAMOZXISYU65HFNNXHOMUY/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":"2026-08-09T15:21:27Z","links":{"resolver":"https://pith.science/pith/AMROHAMOZXISYU65HFNNXHOMUY","bundle":"https://pith.science/pith/AMROHAMOZXISYU65HFNNXHOMUY/bundle.json","state":"https://pith.science/pith/AMROHAMOZXISYU65HFNNXHOMUY/state.json","well_known_bundle":"https://pith.science/.well-known/pith/AMROHAMOZXISYU65HFNNXHOMUY/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:AMROHAMOZXISYU65HFNNXHOMUY","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":"0dccc61c6d0649a8a2ec94dd0e46c0a7a41fb53916af18f7b42f811f3d874c54","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-06-13T01:05:22Z","title_canon_sha256":"97a1ad1c4f626549e32778ffa4e8068b42cdde735891b461467be4a123b4b695"},"schema_version":"1.0","source":{"id":"2406.08725","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.08725","created_at":"2026-07-05T08:31:22Z"},{"alias_kind":"arxiv_version","alias_value":"2406.08725v1","created_at":"2026-07-05T08:31:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.08725","created_at":"2026-07-05T08:31:22Z"},{"alias_kind":"pith_short_12","alias_value":"AMROHAMOZXIS","created_at":"2026-07-05T08:31:22Z"},{"alias_kind":"pith_short_16","alias_value":"AMROHAMOZXISYU65","created_at":"2026-07-05T08:31:22Z"},{"alias_kind":"pith_short_8","alias_value":"AMROHAMO","created_at":"2026-07-05T08:31:22Z"}],"graph_snapshots":[{"event_id":"sha256:3a032d154e486417a9cbd886934bcb7ca9e76299a5abde247288546219c301d2","target":"graph","created_at":"2026-07-05T08:31:22Z","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/2406.08725/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Modern large language model (LLM) developers typically conduct a safety alignment to prevent an LLM from generating unethical or harmful content. Recent studies have discovered that the safety alignment of LLMs can be bypassed by jailbreaking prompts. These prompts are designed to create specific conversation scenarios with a harmful question embedded. Querying an LLM with such prompts can mislead the model into responding to the harmful question. The stochastic and random nature of existing genetic methods largely limits the effectiveness and efficiency of state-of-the-art (SOTA) jailbreaking","authors_text":"Lu Yan, Wenbo Guo, Xiangyu Zhang, Xuan Chen, Yunshu Mao, Yuzhou Nie","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-06-13T01:05:22Z","title":"RL-JACK: Reinforcement Learning-powered Black-box Jailbreaking Attack against LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.08725","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:c7683fd69df233227b18ed6f74a40b6429db0c8ddfac2fb175b868ff9c4cbba3","target":"record","created_at":"2026-07-05T08:31:22Z","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":"0dccc61c6d0649a8a2ec94dd0e46c0a7a41fb53916af18f7b42f811f3d874c54","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-06-13T01:05:22Z","title_canon_sha256":"97a1ad1c4f626549e32778ffa4e8068b42cdde735891b461467be4a123b4b695"},"schema_version":"1.0","source":{"id":"2406.08725","kind":"arxiv","version":1}},"canonical_sha256":"0322e3818ecdd12c53dd395adb9dcca631ba002dcb525a5eeccd0c315151a244","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0322e3818ecdd12c53dd395adb9dcca631ba002dcb525a5eeccd0c315151a244","first_computed_at":"2026-07-05T08:31:22.244658Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:31:22.244658Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"arl9lWDO5XGm/I1w3xyuu++mPtLXVaTe2Sb3nGbX9mtnBvuhFaUyeI2861ugVBrIpDbrZgD6JHUoqlFmo/SOAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T08:31:22.245119Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.08725","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c7683fd69df233227b18ed6f74a40b6429db0c8ddfac2fb175b868ff9c4cbba3","sha256:3a032d154e486417a9cbd886934bcb7ca9e76299a5abde247288546219c301d2"],"state_sha256":"c89737844dbfb4ad67a680edf374a5b4541aa345563d227a61954b806f857e9f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"eWzTJn0hEfwzVnwggmlL6enXv62K+Zfjtigm41TXnfe7FW3QtuQEhK2G0Uruc/qLsvP2vSFC7+ENSLc/maGkBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T15:21:27.923418Z","bundle_sha256":"e18c8229146b37b03f993904b4d6c60fdd6ae7e0f6721348572f620925af4161"}}