{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:34SA2LA5ZRO6NVMZ4OYHTAFCDL","short_pith_number":"pith:34SA2LA5","canonical_record":{"source":{"id":"2408.01605","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-08-02T23:47:27Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"ce7f6708417735867eeb794e409f43942e53de4b1ffae67f569a76bd15a34198","abstract_canon_sha256":"4997b1efa3e3bc3cd79ff62df672fceab818c32a1c3c7b84c1e82f58ea17fc2a"},"schema_version":"1.0"},"canonical_sha256":"df240d2c1dcc5de6d599e3b07980a21af197b5d7e216443778fc25cabda099aa","source":{"kind":"arxiv","id":"2408.01605","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2408.01605","created_at":"2026-07-05T09:04:11Z"},{"alias_kind":"arxiv_version","alias_value":"2408.01605v2","created_at":"2026-07-05T09:04:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.01605","created_at":"2026-07-05T09:04:11Z"},{"alias_kind":"pith_short_12","alias_value":"34SA2LA5ZRO6","created_at":"2026-07-05T09:04:11Z"},{"alias_kind":"pith_short_16","alias_value":"34SA2LA5ZRO6NVMZ","created_at":"2026-07-05T09:04:11Z"},{"alias_kind":"pith_short_8","alias_value":"34SA2LA5","created_at":"2026-07-05T09:04:11Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:34SA2LA5ZRO6NVMZ4OYHTAFCDL","target":"record","payload":{"canonical_record":{"source":{"id":"2408.01605","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-08-02T23:47:27Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"ce7f6708417735867eeb794e409f43942e53de4b1ffae67f569a76bd15a34198","abstract_canon_sha256":"4997b1efa3e3bc3cd79ff62df672fceab818c32a1c3c7b84c1e82f58ea17fc2a"},"schema_version":"1.0"},"canonical_sha256":"df240d2c1dcc5de6d599e3b07980a21af197b5d7e216443778fc25cabda099aa","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:04:11.457136Z","signature_b64":"Lw1a/YkOJUaFvmmcwENzAerVgxUSXJcUCiF0DsEDIW/YJIPEwdnBq7v9N/S53zEyOByHRkHlE3sJe68y0g16Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"df240d2c1dcc5de6d599e3b07980a21af197b5d7e216443778fc25cabda099aa","last_reissued_at":"2026-07-05T09:04:11.456610Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:04:11.456610Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2408.01605","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":"2026-07-05T09:04:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6rfztKl5PfeeAqfq86CMJp0KuN/hYOpSrtHO4/PwTA8pBzoOtqprsg1TpMYgP8arICAQWiZMEULLQXZhrc1eDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-11T21:01:43.380746Z"},"content_sha256":"eb0c3fd50c0371d7296cd5475850cefba7874b372e729baf71a259675efc0984","schema_version":"1.0","event_id":"sha256:eb0c3fd50c0371d7296cd5475850cefba7874b372e729baf71a259675efc0984"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:34SA2LA5ZRO6NVMZ4OYHTAFCDL","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"CYBERSECEVAL 3: Advancing the Evaluation of Cybersecurity Risks and Capabilities in Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CR","authors_text":"Cyrus Nikolaidis, Daniel Song, David Molnar, James Crnkovich, Jayson Grace, Joshua Saxe, Manish Bhatt, Sahana Chennabasappa, Shengye Wan, Spencer Whitman, Stephanie Ding, Vlad Ionescu, Yue Li","submitted_at":"2024-08-02T23:47:27Z","abstract_excerpt":"We are releasing a new suite of security benchmarks for LLMs, CYBERSECEVAL 3, to continue the conversation on empirically measuring LLM cybersecurity risks and capabilities. CYBERSECEVAL 3 assesses 8 different risks across two broad categories: risk to third parties, and risk to application developers and end users. Compared to previous work, we add new areas focused on offensive security capabilities: automated social engineering, scaling manual offensive cyber operations, and autonomous offensive cyber operations. In this paper we discuss applying these benchmarks to the Llama 3 models and a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.01605","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/2408.01605/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":"2026-07-05T09:04:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ljPR2M8Jgf+pdKuLYt7ghlUotxA33PksW5ZNQUQfKxPI8Jgg14TcbHFV4cMuFkeVrPvXnyS48AlZTJDWI3uACA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-11T21:01:43.381125Z"},"content_sha256":"9f9faa7d5eaccde30a5736d3756abbc46790b413799c10e7bc4bcee01df7a397","schema_version":"1.0","event_id":"sha256:9f9faa7d5eaccde30a5736d3756abbc46790b413799c10e7bc4bcee01df7a397"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/34SA2LA5ZRO6NVMZ4OYHTAFCDL/bundle.json","state_url":"https://pith.science/pith/34SA2LA5ZRO6NVMZ4OYHTAFCDL/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/34SA2LA5ZRO6NVMZ4OYHTAFCDL/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":"2026-07-11T21:01:43Z","links":{"resolver":"https://pith.science/pith/34SA2LA5ZRO6NVMZ4OYHTAFCDL","bundle":"https://pith.science/pith/34SA2LA5ZRO6NVMZ4OYHTAFCDL/bundle.json","state":"https://pith.science/pith/34SA2LA5ZRO6NVMZ4OYHTAFCDL/state.json","well_known_bundle":"https://pith.science/.well-known/pith/34SA2LA5ZRO6NVMZ4OYHTAFCDL/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:34SA2LA5ZRO6NVMZ4OYHTAFCDL","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":"4997b1efa3e3bc3cd79ff62df672fceab818c32a1c3c7b84c1e82f58ea17fc2a","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-08-02T23:47:27Z","title_canon_sha256":"ce7f6708417735867eeb794e409f43942e53de4b1ffae67f569a76bd15a34198"},"schema_version":"1.0","source":{"id":"2408.01605","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2408.01605","created_at":"2026-07-05T09:04:11Z"},{"alias_kind":"arxiv_version","alias_value":"2408.01605v2","created_at":"2026-07-05T09:04:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.01605","created_at":"2026-07-05T09:04:11Z"},{"alias_kind":"pith_short_12","alias_value":"34SA2LA5ZRO6","created_at":"2026-07-05T09:04:11Z"},{"alias_kind":"pith_short_16","alias_value":"34SA2LA5ZRO6NVMZ","created_at":"2026-07-05T09:04:11Z"},{"alias_kind":"pith_short_8","alias_value":"34SA2LA5","created_at":"2026-07-05T09:04:11Z"}],"graph_snapshots":[{"event_id":"sha256:9f9faa7d5eaccde30a5736d3756abbc46790b413799c10e7bc4bcee01df7a397","target":"graph","created_at":"2026-07-05T09:04:11Z","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/2408.01605/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We are releasing a new suite of security benchmarks for LLMs, CYBERSECEVAL 3, to continue the conversation on empirically measuring LLM cybersecurity risks and capabilities. CYBERSECEVAL 3 assesses 8 different risks across two broad categories: risk to third parties, and risk to application developers and end users. Compared to previous work, we add new areas focused on offensive security capabilities: automated social engineering, scaling manual offensive cyber operations, and autonomous offensive cyber operations. In this paper we discuss applying these benchmarks to the Llama 3 models and a","authors_text":"Cyrus Nikolaidis, Daniel Song, David Molnar, James Crnkovich, Jayson Grace, Joshua Saxe, Manish Bhatt, Sahana Chennabasappa, Shengye Wan, Spencer Whitman, Stephanie Ding, Vlad Ionescu, Yue Li","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-08-02T23:47:27Z","title":"CYBERSECEVAL 3: Advancing the Evaluation of Cybersecurity Risks and Capabilities in Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.01605","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:eb0c3fd50c0371d7296cd5475850cefba7874b372e729baf71a259675efc0984","target":"record","created_at":"2026-07-05T09:04:11Z","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":"4997b1efa3e3bc3cd79ff62df672fceab818c32a1c3c7b84c1e82f58ea17fc2a","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-08-02T23:47:27Z","title_canon_sha256":"ce7f6708417735867eeb794e409f43942e53de4b1ffae67f569a76bd15a34198"},"schema_version":"1.0","source":{"id":"2408.01605","kind":"arxiv","version":2}},"canonical_sha256":"df240d2c1dcc5de6d599e3b07980a21af197b5d7e216443778fc25cabda099aa","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"df240d2c1dcc5de6d599e3b07980a21af197b5d7e216443778fc25cabda099aa","first_computed_at":"2026-07-05T09:04:11.456610Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:04:11.456610Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Lw1a/YkOJUaFvmmcwENzAerVgxUSXJcUCiF0DsEDIW/YJIPEwdnBq7v9N/S53zEyOByHRkHlE3sJe68y0g16Ag==","signature_status":"signed_v1","signed_at":"2026-07-05T09:04:11.457136Z","signed_message":"canonical_sha256_bytes"},"source_id":"2408.01605","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:eb0c3fd50c0371d7296cd5475850cefba7874b372e729baf71a259675efc0984","sha256:9f9faa7d5eaccde30a5736d3756abbc46790b413799c10e7bc4bcee01df7a397"],"state_sha256":"e1a6141e42052e1c82051db1b5c7f2e7da25e352bc677e074bc95d43418eae64"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"uJau0lfG+PrGGcfS4AAZhMQO6EHqgwKa6tHnuTx2dhLKVpqMkL4+R9UfmzaCZvWkVTECI6/Xgm/z1DSIiQn2CA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-11T21:01:43.383380Z","bundle_sha256":"01c73bf1fcc1d48cdf4ceb624b5a0ebfe5bdbd97955fd1598a5396a1cded4f06"}}