{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:A5SBQQMMO5JWGP7YIZCQ4V2AZF","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":"ffdaac23185ef7b6ecb9f91ae063f92b854da3aee9f636a76227e7ec92b32a6a","cross_cats_sorted":["cs.AI","cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-04-13T20:21:08Z","title_canon_sha256":"5ce9371552e9b22956213a1149a670c7e114aa79c27f5c53983de2af56102a38"},"schema_version":"1.0","source":{"id":"2504.09712","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.09712","created_at":"2026-07-05T11:13:24Z"},{"alias_kind":"arxiv_version","alias_value":"2504.09712v2","created_at":"2026-07-05T11:13:24Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.09712","created_at":"2026-07-05T11:13:24Z"},{"alias_kind":"pith_short_12","alias_value":"A5SBQQMMO5JW","created_at":"2026-07-05T11:13:24Z"},{"alias_kind":"pith_short_16","alias_value":"A5SBQQMMO5JWGP7Y","created_at":"2026-07-05T11:13:24Z"},{"alias_kind":"pith_short_8","alias_value":"A5SBQQMM","created_at":"2026-07-05T11:13:24Z"}],"graph_snapshots":[{"event_id":"sha256:8ec954e76901d114567c7834606fba818eb4f7275ea3a9bf01b8a6229c6749a8","target":"graph","created_at":"2026-07-05T11:13:24Z","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/2504.09712/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"LLM jailbreaks are a widespread safety challenge. Given this problem has not yet been tractable, we suggest targeting a key failure mechanism: the failure of safety to generalize across semantically equivalent inputs. We further focus the target by requiring desirable tractability properties of attacks to study: explainability, transferability between models, and transferability between goals. We perform red-teaming within this framework by uncovering new vulnerabilities to multi-turn, multi-image, and translation-based attacks. These attacks are semantically equivalent by our design to their ","authors_text":"Ethan Kosak-Hine, George Ingebretsen, Jason Zhang, Julius Broomfield, Kellin Pelrine, Reihaneh Iranmanesh, Reihaneh Rabbany, Sara Pieri, Tia Nasir, Tom Gibbs","cross_cats":["cs.AI","cs.CV"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-04-13T20:21:08Z","title":"The Structural Safety Generalization Problem"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.09712","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:c771555af8948796d0767bef550c809f3265e578dc667c059b88fa07ab513b0b","target":"record","created_at":"2026-07-05T11:13:24Z","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":"ffdaac23185ef7b6ecb9f91ae063f92b854da3aee9f636a76227e7ec92b32a6a","cross_cats_sorted":["cs.AI","cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-04-13T20:21:08Z","title_canon_sha256":"5ce9371552e9b22956213a1149a670c7e114aa79c27f5c53983de2af56102a38"},"schema_version":"1.0","source":{"id":"2504.09712","kind":"arxiv","version":2}},"canonical_sha256":"076418418c7753633ff846450e5740c9516e1df80644930bdc745699de5b1c0b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"076418418c7753633ff846450e5740c9516e1df80644930bdc745699de5b1c0b","first_computed_at":"2026-07-05T11:13:24.131366Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:13:24.131366Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"UDs4UcebUk1DzsF7C+TGLX46YPxWxYJnQ4+8QnQBvag5canp9F2x8hUe7B2cgnQookGnwJeeli2x4dqDPBhfAw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:13:24.131850Z","signed_message":"canonical_sha256_bytes"},"source_id":"2504.09712","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c771555af8948796d0767bef550c809f3265e578dc667c059b88fa07ab513b0b","sha256:8ec954e76901d114567c7834606fba818eb4f7275ea3a9bf01b8a6229c6749a8"],"state_sha256":"22f759135830caaca2e0bb649ab0b0a86415b605ed4b2efc024888836af8d702"}