{"paper":{"title":"Jailbreaking Large Language Models Against Moderation Guardrails via Cipher Characters","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.CV","cs.LG"],"primary_cat":"cs.CR","authors_text":"Andy Zhou, Haibo Jin, Haohan Wang, Joe D. Menke","submitted_at":"2024-05-30T18:38:36Z","abstract_excerpt":"Large Language Models (LLMs) are typically harmless but remain vulnerable to carefully crafted prompts known as ``jailbreaks'', which can bypass protective measures and induce harmful behavior. Recent advancements in LLMs have incorporated moderation guardrails that can filter outputs, which trigger processing errors for certain malicious questions. Existing red-teaming benchmarks often neglect to include questions that trigger moderation guardrails, making it difficult to evaluate jailbreak effectiveness. To address this issue, we introduce JAMBench, a harmful behavior benchmark designed to t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.20413","kind":"arxiv","version":1},"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/2405.20413/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"}