{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:BGGXTDTBMDHVTJJXNLGJR5B7HA","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":"833c057ed1ec5a9ba33c675b8250cb4b1dcd07e4e010b03b80f9a406aa663b92","cross_cats_sorted":["cs.AI","cs.CL","cs.CR"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-10-21T17:27:01Z","title_canon_sha256":"442dbd2a9ed8c0e2d7ad25934cf782699ce844fef6c04d78044bdc5e55c76c52"},"schema_version":"1.0","source":{"id":"2410.16222","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.16222","created_at":"2026-07-05T11:19:49Z"},{"alias_kind":"arxiv_version","alias_value":"2410.16222v2","created_at":"2026-07-05T11:19:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.16222","created_at":"2026-07-05T11:19:49Z"},{"alias_kind":"pith_short_12","alias_value":"BGGXTDTBMDHV","created_at":"2026-07-05T11:19:49Z"},{"alias_kind":"pith_short_16","alias_value":"BGGXTDTBMDHVTJJX","created_at":"2026-07-05T11:19:49Z"},{"alias_kind":"pith_short_8","alias_value":"BGGXTDTB","created_at":"2026-07-05T11:19:49Z"}],"graph_snapshots":[{"event_id":"sha256:28e547d6b1a709ff46aafbd9876be6bcf54394d8c855dedc3aaf71297a0ffecc","target":"graph","created_at":"2026-07-05T11:19:49Z","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/2410.16222/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"A plethora of jailbreaking attacks have been proposed to obtain harmful responses from safety-tuned LLMs. These methods largely succeed in coercing the target output in their original settings, but their attacks vary substantially in fluency and computational effort. In this work, we propose a unified threat model for the principled comparison of these methods. Our threat model checks if a given jailbreak is likely to occur in the distribution of text. For this, we build an N-gram language model on 1T tokens, which, unlike model-based perplexity, allows for an LLM-agnostic, nonparametric, and ","authors_text":"Alexander Panfilov, Jonas Geiping, Matthias Hein, Vaclav Voracek, Valentyn Boreiko","cross_cats":["cs.AI","cs.CL","cs.CR"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-10-21T17:27:01Z","title":"An Interpretable N-gram Perplexity Threat Model for Large Language Model Jailbreaks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.16222","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:b1ea7c48f49ea60ac37e1971802a1527ed04f620192e50ac37d3de00b16760ee","target":"record","created_at":"2026-07-05T11:19:49Z","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":"833c057ed1ec5a9ba33c675b8250cb4b1dcd07e4e010b03b80f9a406aa663b92","cross_cats_sorted":["cs.AI","cs.CL","cs.CR"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-10-21T17:27:01Z","title_canon_sha256":"442dbd2a9ed8c0e2d7ad25934cf782699ce844fef6c04d78044bdc5e55c76c52"},"schema_version":"1.0","source":{"id":"2410.16222","kind":"arxiv","version":2}},"canonical_sha256":"098d798e6160cf59a5376acc98f43f381f0ae2f3c517a0ab22ba687edc681f24","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"098d798e6160cf59a5376acc98f43f381f0ae2f3c517a0ab22ba687edc681f24","first_computed_at":"2026-07-05T11:19:49.839608Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:19:49.839608Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"2phQ9Ice/NYNyBzuohWI6aKipJMWKq6xwxe7Rjuq3n35OLI3oeUYBhA2dzI5rRYbM75OiS+7C4SWInLRdscfCw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:19:49.840048Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.16222","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b1ea7c48f49ea60ac37e1971802a1527ed04f620192e50ac37d3de00b16760ee","sha256:28e547d6b1a709ff46aafbd9876be6bcf54394d8c855dedc3aaf71297a0ffecc"],"state_sha256":"7b3ccc0e1accea10eb400a26f819f383a4cd35e4fcd5e2dc87c45476b420da58"}