{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:OTA2I4NJZCJ5VTPATZ4RIQKDKQ","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":"70ca0f360eefb9ec008195351fa2b95f0e97c663faeaf65d971e45dfe821a186","cross_cats_sorted":["cs.CV","cs.NE"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-09-04T08:54:20Z","title_canon_sha256":"982999b78d4c52fd6baeceb6496e211ca2992489410463b3afa8ee954fa154df"},"schema_version":"1.0","source":{"id":"2309.01446","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2309.01446","created_at":"2026-07-05T08:51:43Z"},{"alias_kind":"arxiv_version","alias_value":"2309.01446v4","created_at":"2026-07-05T08:51:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.01446","created_at":"2026-07-05T08:51:43Z"},{"alias_kind":"pith_short_12","alias_value":"OTA2I4NJZCJ5","created_at":"2026-07-05T08:51:43Z"},{"alias_kind":"pith_short_16","alias_value":"OTA2I4NJZCJ5VTPA","created_at":"2026-07-05T08:51:43Z"},{"alias_kind":"pith_short_8","alias_value":"OTA2I4NJ","created_at":"2026-07-05T08:51:43Z"}],"graph_snapshots":[{"event_id":"sha256:acd2c4658612e21802cd89e9105ce5a8409503934ae1276259231780e867a6b1","target":"graph","created_at":"2026-07-05T08:51:43Z","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/2309.01446/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs), designed to provide helpful and safe responses, often rely on alignment techniques to align with user intent and social guidelines. Unfortunately, this alignment can be exploited by malicious actors seeking to manipulate an LLM's outputs for unintended purposes. In this paper we introduce a novel approach that employs a genetic algorithm (GA) to manipulate LLMs when model architecture and parameters are inaccessible. The GA attack works by optimizing a universal adversarial prompt that -- when combined with a user's query -- disrupts the attacked model's alignment","authors_text":"Moshe Sipper, Raz Lapid, Ron Langberg","cross_cats":["cs.CV","cs.NE"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-09-04T08:54:20Z","title":"Open Sesame! Universal Black Box Jailbreaking of Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.01446","kind":"arxiv","version":4},"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":"sha256:7c1bfcdddce880773333b20ac01698363dd164239fd93e4931169323a190b72c","target":"record","created_at":"2026-07-05T08:51:43Z","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":"70ca0f360eefb9ec008195351fa2b95f0e97c663faeaf65d971e45dfe821a186","cross_cats_sorted":["cs.CV","cs.NE"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-09-04T08:54:20Z","title_canon_sha256":"982999b78d4c52fd6baeceb6496e211ca2992489410463b3afa8ee954fa154df"},"schema_version":"1.0","source":{"id":"2309.01446","kind":"arxiv","version":4}},"canonical_sha256":"74c1a471a9c893dacde09e79144143542eb0f941067c13867ca14ebcfbc75384","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"74c1a471a9c893dacde09e79144143542eb0f941067c13867ca14ebcfbc75384","first_computed_at":"2026-07-05T08:51:43.352840Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:51:43.352840Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"1ybkDEWL708xRaW33ARbb0Ri8JWrcDpKXI4VymMNKZlUM7iliyg8qibQTl2+cY4MMOdt5t7wFowDFzTJXDDmAw==","signature_status":"signed_v1","signed_at":"2026-07-05T08:51:43.353311Z","signed_message":"canonical_sha256_bytes"},"source_id":"2309.01446","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7c1bfcdddce880773333b20ac01698363dd164239fd93e4931169323a190b72c","sha256:acd2c4658612e21802cd89e9105ce5a8409503934ae1276259231780e867a6b1"],"state_sha256":"6a3be5609cc29d0cdfb066168cee58db1b557d2f646ba1d5d7eb615cc1915087"}