{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:NFTK5Q66KY4ZAEMAW5SORBJSUH","short_pith_number":"pith:NFTK5Q66","schema_version":"1.0","canonical_sha256":"6966aec3de5639901180b764e88532a1d93f90251ea4eba5384d503cf335982f","source":{"kind":"arxiv","id":"2606.26936","version":1},"attestation_state":"computed","paper":{"title":"Jailbreaking for the Average Jane: Choosing Optimal Jailbreaks via Bandit Algorithms for Automatically Enhanced Queries","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.CR","authors_text":"Arjun Bhagoji, Arpit Agarwal, Prarabdh Shukla, Ritik, Suhas Rao","submitted_at":"2026-06-25T12:11:28Z","abstract_excerpt":"With a profusion of jailbreaks for LLMs now widely known, a growing concern is that non-expert malicious actors (\"the average Jane\") could elicit actionable responses to malicious requests. In this work, we examine whether this concern is justified. A non-expert malicious actor requires two ingredients for a successful attack: a powerful jailbreak for their target model, acting on an effective malicious query. For the former, we propose a novel attack strategy based on the multi-armed bandit framework. This allows efficient online learning of the optimal jailbreak from a large choice set via n"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2606.26936","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2026-06-25T12:11:28Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"f67fef4ef52ed522ceffc2e9b76a6797855b814b07f277d97678cd10d503000f","abstract_canon_sha256":"e6765b068e1b89869a95af0ec92fe3c52994a38e7132409e7461e9e406e061c4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-26T01:16:04.562623Z","signature_b64":"iS+ZABB3AiLZkDqgGhzeNK3mShX0XiUeaWyCNV17lSyIhhfs1NfhqgkSkr8Kxsnnt9/ZZw0TJmGiOccn7/thCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6966aec3de5639901180b764e88532a1d93f90251ea4eba5384d503cf335982f","last_reissued_at":"2026-06-26T01:16:04.562165Z","signature_status":"signed_v1","first_computed_at":"2026-06-26T01:16:04.562165Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Jailbreaking for the Average Jane: Choosing Optimal Jailbreaks via Bandit Algorithms for Automatically Enhanced Queries","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.CR","authors_text":"Arjun Bhagoji, Arpit Agarwal, Prarabdh Shukla, Ritik, Suhas Rao","submitted_at":"2026-06-25T12:11:28Z","abstract_excerpt":"With a profusion of jailbreaks for LLMs now widely known, a growing concern is that non-expert malicious actors (\"the average Jane\") could elicit actionable responses to malicious requests. In this work, we examine whether this concern is justified. A non-expert malicious actor requires two ingredients for a successful attack: a powerful jailbreak for their target model, acting on an effective malicious query. For the former, we propose a novel attack strategy based on the multi-armed bandit framework. This allows efficient online learning of the optimal jailbreak from a large choice set via n"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.26936","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/2606.26936/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2606.26936","created_at":"2026-06-26T01:16:04.562225+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.26936v1","created_at":"2026-06-26T01:16:04.562225+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.26936","created_at":"2026-06-26T01:16:04.562225+00:00"},{"alias_kind":"pith_short_12","alias_value":"NFTK5Q66KY4Z","created_at":"2026-06-26T01:16:04.562225+00:00"},{"alias_kind":"pith_short_16","alias_value":"NFTK5Q66KY4ZAEMA","created_at":"2026-06-26T01:16:04.562225+00:00"},{"alias_kind":"pith_short_8","alias_value":"NFTK5Q66","created_at":"2026-06-26T01:16:04.562225+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NFTK5Q66KY4ZAEMAW5SORBJSUH","json":"https://pith.science/pith/NFTK5Q66KY4ZAEMAW5SORBJSUH.json","graph_json":"https://pith.science/api/pith-number/NFTK5Q66KY4ZAEMAW5SORBJSUH/graph.json","events_json":"https://pith.science/api/pith-number/NFTK5Q66KY4ZAEMAW5SORBJSUH/events.json","paper":"https://pith.science/paper/NFTK5Q66"},"agent_actions":{"view_html":"https://pith.science/pith/NFTK5Q66KY4ZAEMAW5SORBJSUH","download_json":"https://pith.science/pith/NFTK5Q66KY4ZAEMAW5SORBJSUH.json","view_paper":"https://pith.science/paper/NFTK5Q66","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.26936&json=true","fetch_graph":"https://pith.science/api/pith-number/NFTK5Q66KY4ZAEMAW5SORBJSUH/graph.json","fetch_events":"https://pith.science/api/pith-number/NFTK5Q66KY4ZAEMAW5SORBJSUH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NFTK5Q66KY4ZAEMAW5SORBJSUH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NFTK5Q66KY4ZAEMAW5SORBJSUH/action/storage_attestation","attest_author":"https://pith.science/pith/NFTK5Q66KY4ZAEMAW5SORBJSUH/action/author_attestation","sign_citation":"https://pith.science/pith/NFTK5Q66KY4ZAEMAW5SORBJSUH/action/citation_signature","submit_replication":"https://pith.science/pith/NFTK5Q66KY4ZAEMAW5SORBJSUH/action/replication_record"}},"created_at":"2026-06-26T01:16:04.562225+00:00","updated_at":"2026-06-26T01:16:04.562225+00:00"}