{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:OGBTINMGWGXAIV67SZUJQVALR5","short_pith_number":"pith:OGBTINMG","canonical_record":{"source":{"id":"2404.13161","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-04-19T20:11:12Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"da5682b79c6e4a5430e7a3cfbe661b6629bdddfcbe1ae8637b6f162d2ff1d54e","abstract_canon_sha256":"f60d3294ba59e41a5fc5c02d6fbc6cca8e8c533bfc81adad8f8596e628d73b95"},"schema_version":"1.0"},"canonical_sha256":"7183343586b1ae0457df966898540b8f78e0cd0cb032fecf6111b5d38542fcbe","source":{"kind":"arxiv","id":"2404.13161","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.13161","created_at":"2026-07-05T08:10:28Z"},{"alias_kind":"arxiv_version","alias_value":"2404.13161v1","created_at":"2026-07-05T08:10:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.13161","created_at":"2026-07-05T08:10:28Z"},{"alias_kind":"pith_short_12","alias_value":"OGBTINMGWGXA","created_at":"2026-07-05T08:10:28Z"},{"alias_kind":"pith_short_16","alias_value":"OGBTINMGWGXAIV67","created_at":"2026-07-05T08:10:28Z"},{"alias_kind":"pith_short_8","alias_value":"OGBTINMG","created_at":"2026-07-05T08:10:28Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:OGBTINMGWGXAIV67SZUJQVALR5","target":"record","payload":{"canonical_record":{"source":{"id":"2404.13161","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-04-19T20:11:12Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"da5682b79c6e4a5430e7a3cfbe661b6629bdddfcbe1ae8637b6f162d2ff1d54e","abstract_canon_sha256":"f60d3294ba59e41a5fc5c02d6fbc6cca8e8c533bfc81adad8f8596e628d73b95"},"schema_version":"1.0"},"canonical_sha256":"7183343586b1ae0457df966898540b8f78e0cd0cb032fecf6111b5d38542fcbe","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:10:28.295192Z","signature_b64":"VmbDRQ/rPosWG3SBxcIKlBmdGASxxdTlzvT5gFefNjeLGWcWs0BIS3t1YBebtJbzw0xgzV6Duj54dGRPT/opBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7183343586b1ae0457df966898540b8f78e0cd0cb032fecf6111b5d38542fcbe","last_reissued_at":"2026-07-05T08:10:28.294692Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:10:28.294692Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2404.13161","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":"2026-07-05T08:10:28Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"InkxXwcKDdibHVBar/0cw7IlS9vC86VN+j4Ci7vB+Zp3Xk3g0Hu946jzdevzJayvIecXRKvrnPzswfvBiRGwAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-01T05:37:14.278376Z"},"content_sha256":"ac9df86ff229609cbd8eb93eeb39ea9d47b3cad8fc7fbe71e5111876e1c3c5a1","schema_version":"1.0","event_id":"sha256:ac9df86ff229609cbd8eb93eeb39ea9d47b3cad8fc7fbe71e5111876e1c3c5a1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:OGBTINMGWGXAIV67SZUJQVALR5","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"CyberSecEval 2: A Wide-Ranging Cybersecurity Evaluation Suite for Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CR","authors_text":"Cornelius Aschermann, Cyrus Nikolaidis, Daniel Song, David Molnar, Dhaval Kapil, Faizan Ahmad, Joshua Saxe, Manish Bhatt, Sahana Chennabasappa, Shengye Wan, Spencer Whitman, Yaohui Chen, Yue Li","submitted_at":"2024-04-19T20:11:12Z","abstract_excerpt":"Large language models (LLMs) introduce new security risks, but there are few comprehensive evaluation suites to measure and reduce these risks. We present BenchmarkName, a novel benchmark to quantify LLM security risks and capabilities. We introduce two new areas for testing: prompt injection and code interpreter abuse. We evaluated multiple state-of-the-art (SOTA) LLMs, including GPT-4, Mistral, Meta Llama 3 70B-Instruct, and Code Llama. Our results show that conditioning away risk of attack remains an unsolved problem; for example, all tested models showed between 26% and 41% successful prom"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.13161","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/2404.13161/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":"2026-07-05T08:10:28Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IPLCxL1St+oXfwqvEYMwS8oPip8eif/LuobzmuLO1DykmAGgnC5nA9bsfC7FjgTBaPNC5da2gP0Of+eMRuqTCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-01T05:37:14.279055Z"},"content_sha256":"c28e477b122eaa6d6b1f7a60a8f3dac8681d5c15d373f7e5d6ca45bfe424d617","schema_version":"1.0","event_id":"sha256:c28e477b122eaa6d6b1f7a60a8f3dac8681d5c15d373f7e5d6ca45bfe424d617"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/OGBTINMGWGXAIV67SZUJQVALR5/bundle.json","state_url":"https://pith.science/pith/OGBTINMGWGXAIV67SZUJQVALR5/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/OGBTINMGWGXAIV67SZUJQVALR5/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":"2026-08-01T05:37:14Z","links":{"resolver":"https://pith.science/pith/OGBTINMGWGXAIV67SZUJQVALR5","bundle":"https://pith.science/pith/OGBTINMGWGXAIV67SZUJQVALR5/bundle.json","state":"https://pith.science/pith/OGBTINMGWGXAIV67SZUJQVALR5/state.json","well_known_bundle":"https://pith.science/.well-known/pith/OGBTINMGWGXAIV67SZUJQVALR5/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:OGBTINMGWGXAIV67SZUJQVALR5","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":"f60d3294ba59e41a5fc5c02d6fbc6cca8e8c533bfc81adad8f8596e628d73b95","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-04-19T20:11:12Z","title_canon_sha256":"da5682b79c6e4a5430e7a3cfbe661b6629bdddfcbe1ae8637b6f162d2ff1d54e"},"schema_version":"1.0","source":{"id":"2404.13161","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.13161","created_at":"2026-07-05T08:10:28Z"},{"alias_kind":"arxiv_version","alias_value":"2404.13161v1","created_at":"2026-07-05T08:10:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.13161","created_at":"2026-07-05T08:10:28Z"},{"alias_kind":"pith_short_12","alias_value":"OGBTINMGWGXA","created_at":"2026-07-05T08:10:28Z"},{"alias_kind":"pith_short_16","alias_value":"OGBTINMGWGXAIV67","created_at":"2026-07-05T08:10:28Z"},{"alias_kind":"pith_short_8","alias_value":"OGBTINMG","created_at":"2026-07-05T08:10:28Z"}],"graph_snapshots":[{"event_id":"sha256:c28e477b122eaa6d6b1f7a60a8f3dac8681d5c15d373f7e5d6ca45bfe424d617","target":"graph","created_at":"2026-07-05T08:10:28Z","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/2404.13161/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) introduce new security risks, but there are few comprehensive evaluation suites to measure and reduce these risks. We present BenchmarkName, a novel benchmark to quantify LLM security risks and capabilities. We introduce two new areas for testing: prompt injection and code interpreter abuse. We evaluated multiple state-of-the-art (SOTA) LLMs, including GPT-4, Mistral, Meta Llama 3 70B-Instruct, and Code Llama. Our results show that conditioning away risk of attack remains an unsolved problem; for example, all tested models showed between 26% and 41% successful prom","authors_text":"Cornelius Aschermann, Cyrus Nikolaidis, Daniel Song, David Molnar, Dhaval Kapil, Faizan Ahmad, Joshua Saxe, Manish Bhatt, Sahana Chennabasappa, Shengye Wan, Spencer Whitman, Yaohui Chen, Yue Li","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-04-19T20:11:12Z","title":"CyberSecEval 2: A Wide-Ranging Cybersecurity Evaluation Suite for Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.13161","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:ac9df86ff229609cbd8eb93eeb39ea9d47b3cad8fc7fbe71e5111876e1c3c5a1","target":"record","created_at":"2026-07-05T08:10:28Z","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":"f60d3294ba59e41a5fc5c02d6fbc6cca8e8c533bfc81adad8f8596e628d73b95","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-04-19T20:11:12Z","title_canon_sha256":"da5682b79c6e4a5430e7a3cfbe661b6629bdddfcbe1ae8637b6f162d2ff1d54e"},"schema_version":"1.0","source":{"id":"2404.13161","kind":"arxiv","version":1}},"canonical_sha256":"7183343586b1ae0457df966898540b8f78e0cd0cb032fecf6111b5d38542fcbe","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7183343586b1ae0457df966898540b8f78e0cd0cb032fecf6111b5d38542fcbe","first_computed_at":"2026-07-05T08:10:28.294692Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:10:28.294692Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"VmbDRQ/rPosWG3SBxcIKlBmdGASxxdTlzvT5gFefNjeLGWcWs0BIS3t1YBebtJbzw0xgzV6Duj54dGRPT/opBA==","signature_status":"signed_v1","signed_at":"2026-07-05T08:10:28.295192Z","signed_message":"canonical_sha256_bytes"},"source_id":"2404.13161","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ac9df86ff229609cbd8eb93eeb39ea9d47b3cad8fc7fbe71e5111876e1c3c5a1","sha256:c28e477b122eaa6d6b1f7a60a8f3dac8681d5c15d373f7e5d6ca45bfe424d617"],"state_sha256":"610c8c82e459d7af281ad3be9c5cb5ca6a06d177b235eea990770c3ca9a04d49"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"jqOefZeAeunnkmIZ6p/9iqIfZOeXEuV9qZymWSnO6rYyz8gfoFb9ZuYc4PPom/rVGhj2RNVjqeQP5DhbBgkfAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-01T05:37:14.287244Z","bundle_sha256":"8ccb387c825c46c358c100b035a8a049e44bb32d83cb9eff2035cd9bb61a7bbe"}}