{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:KMDB4M4SW37Q7AAX3YGUB7Y3DM","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":"786b40a8a1552e7d8707930c2f1bc338f718d2186e85c8698236fcc5f122d21d","cross_cats_sorted":["cs.CL","cs.CR","cs.CY"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-08-27T17:33:30Z","title_canon_sha256":"f934fce144a82232edcee5b21c075bb368e8981af2deb0d44f69211000d19d7e"},"schema_version":"1.0","source":{"id":"2408.15221","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2408.15221","created_at":"2026-07-05T09:02:41Z"},{"alias_kind":"arxiv_version","alias_value":"2408.15221v2","created_at":"2026-07-05T09:02:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.15221","created_at":"2026-07-05T09:02:41Z"},{"alias_kind":"pith_short_12","alias_value":"KMDB4M4SW37Q","created_at":"2026-07-05T09:02:41Z"},{"alias_kind":"pith_short_16","alias_value":"KMDB4M4SW37Q7AAX","created_at":"2026-07-05T09:02:41Z"},{"alias_kind":"pith_short_8","alias_value":"KMDB4M4S","created_at":"2026-07-05T09:02:41Z"}],"graph_snapshots":[{"event_id":"sha256:591371558f531e8d8a27265d61d1f97cc3e9413e1acb7926fd72f458b68345dc","target":"graph","created_at":"2026-07-05T09:02:41Z","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/2408.15221/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent large language model (LLM) defenses have greatly improved models' ability to refuse harmful queries, even when adversarially attacked. However, LLM defenses are primarily evaluated against automated adversarial attacks in a single turn of conversation, an insufficient threat model for real-world malicious use. We demonstrate that multi-turn human jailbreaks uncover significant vulnerabilities, exceeding 70% attack success rate (ASR) on HarmBench against defenses that report single-digit ASRs with automated single-turn attacks. Human jailbreaks also reveal vulnerabilities in machine unle","authors_text":"Cristina Menghini, Hugh Zhang, Ian Steneker, Nathaniel Li, Riley Goodside, Summer Yue, Willow Primack, Zifan Wang, Ziwen Han","cross_cats":["cs.CL","cs.CR","cs.CY"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-08-27T17:33:30Z","title":"LLM Defenses Are Not Robust to Multi-Turn Human Jailbreaks Yet"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.15221","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":"sha256:2c99825dbb31b22f8431f3b4d577e4083997cb722b9c0314a2e9d7aa37ec9cab","target":"record","created_at":"2026-07-05T09:02:41Z","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":"786b40a8a1552e7d8707930c2f1bc338f718d2186e85c8698236fcc5f122d21d","cross_cats_sorted":["cs.CL","cs.CR","cs.CY"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-08-27T17:33:30Z","title_canon_sha256":"f934fce144a82232edcee5b21c075bb368e8981af2deb0d44f69211000d19d7e"},"schema_version":"1.0","source":{"id":"2408.15221","kind":"arxiv","version":2}},"canonical_sha256":"53061e3392b6ff0f8017de0d40ff1b1b202626d6741dcbfdd1a7bade0ad9a3b0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"53061e3392b6ff0f8017de0d40ff1b1b202626d6741dcbfdd1a7bade0ad9a3b0","first_computed_at":"2026-07-05T09:02:41.943898Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:02:41.943898Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"AzsOP4XZ0o8n7Z3H4LA8SZctFj//4oZppWvz3qJAorEABuOqYaxAd6kj9B2iBjU5Hel6Bo956jLAMoOo+qnDDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:02:41.944426Z","signed_message":"canonical_sha256_bytes"},"source_id":"2408.15221","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2c99825dbb31b22f8431f3b4d577e4083997cb722b9c0314a2e9d7aa37ec9cab","sha256:591371558f531e8d8a27265d61d1f97cc3e9413e1acb7926fd72f458b68345dc"],"state_sha256":"73745967ed02713be9fd91d44300c2fe4b38bd500f760ed211305d2a1893800a"}