{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:A5JO4TUXHJP7PRYGKAKBZUGG2W","short_pith_number":"pith:A5JO4TUX","schema_version":"1.0","canonical_sha256":"0752ee4e973a5ff7c70650141cd0c6d58485812aa1915aea5fa2e59ee0d5c70c","source":{"kind":"arxiv","id":"2410.15645","version":2},"attestation_state":"computed","paper":{"title":"Boosting Jailbreak Transferability for Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Hanqing Liu, Huanqian Yan, Lifeng Zhou","submitted_at":"2024-10-21T05:11:19Z","abstract_excerpt":"Large language models have drawn significant attention to the challenge of safe alignment, especially regarding jailbreak attacks that circumvent security measures to produce harmful content. To address the limitations of existing methods like GCG, which perform well in single-model attacks but lack transferability, we propose several enhancements, including a scenario induction template, optimized suffix selection, and the integration of re-suffix attack mechanism to reduce inconsistent outputs. Our approach has shown superior performance in extensive experiments across various benchmarks, ac"},"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":"2410.15645","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-10-21T05:11:19Z","cross_cats_sorted":[],"title_canon_sha256":"7af4d7a31641632a7238c2cd70d2c1e784ae9caa05be458dbea94a93f03261c8","abstract_canon_sha256":"5b870cac5e24ad3257595fc0be840818cedf1b175cc53888cad1954bd0b5d600"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:30:22.502300Z","signature_b64":"eJFw1YgsIoV0tVC4WxK5D7dt1GUrZf2ujx2yNSN9oFNYQoq0zjXF/1P20LSAZQVXcfUkSaYDLyxCIz43RzvDDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0752ee4e973a5ff7c70650141cd0c6d58485812aa1915aea5fa2e59ee0d5c70c","last_reissued_at":"2026-07-05T09:30:22.501852Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:30:22.501852Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Boosting Jailbreak Transferability for Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Hanqing Liu, Huanqian Yan, Lifeng Zhou","submitted_at":"2024-10-21T05:11:19Z","abstract_excerpt":"Large language models have drawn significant attention to the challenge of safe alignment, especially regarding jailbreak attacks that circumvent security measures to produce harmful content. To address the limitations of existing methods like GCG, which perform well in single-model attacks but lack transferability, we propose several enhancements, including a scenario induction template, optimized suffix selection, and the integration of re-suffix attack mechanism to reduce inconsistent outputs. Our approach has shown superior performance in extensive experiments across various benchmarks, ac"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.15645","kind":"arxiv","version":2},"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/2410.15645/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":"2410.15645","created_at":"2026-07-05T09:30:22.501917+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.15645v2","created_at":"2026-07-05T09:30:22.501917+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.15645","created_at":"2026-07-05T09:30:22.501917+00:00"},{"alias_kind":"pith_short_12","alias_value":"A5JO4TUXHJP7","created_at":"2026-07-05T09:30:22.501917+00:00"},{"alias_kind":"pith_short_16","alias_value":"A5JO4TUXHJP7PRYG","created_at":"2026-07-05T09:30:22.501917+00:00"},{"alias_kind":"pith_short_8","alias_value":"A5JO4TUX","created_at":"2026-07-05T09:30:22.501917+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.05498","citing_title":"JailWAM: Jailbreaking World Action Models in Robot Control","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/A5JO4TUXHJP7PRYGKAKBZUGG2W","json":"https://pith.science/pith/A5JO4TUXHJP7PRYGKAKBZUGG2W.json","graph_json":"https://pith.science/api/pith-number/A5JO4TUXHJP7PRYGKAKBZUGG2W/graph.json","events_json":"https://pith.science/api/pith-number/A5JO4TUXHJP7PRYGKAKBZUGG2W/events.json","paper":"https://pith.science/paper/A5JO4TUX"},"agent_actions":{"view_html":"https://pith.science/pith/A5JO4TUXHJP7PRYGKAKBZUGG2W","download_json":"https://pith.science/pith/A5JO4TUXHJP7PRYGKAKBZUGG2W.json","view_paper":"https://pith.science/paper/A5JO4TUX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.15645&json=true","fetch_graph":"https://pith.science/api/pith-number/A5JO4TUXHJP7PRYGKAKBZUGG2W/graph.json","fetch_events":"https://pith.science/api/pith-number/A5JO4TUXHJP7PRYGKAKBZUGG2W/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/A5JO4TUXHJP7PRYGKAKBZUGG2W/action/timestamp_anchor","attest_storage":"https://pith.science/pith/A5JO4TUXHJP7PRYGKAKBZUGG2W/action/storage_attestation","attest_author":"https://pith.science/pith/A5JO4TUXHJP7PRYGKAKBZUGG2W/action/author_attestation","sign_citation":"https://pith.science/pith/A5JO4TUXHJP7PRYGKAKBZUGG2W/action/citation_signature","submit_replication":"https://pith.science/pith/A5JO4TUXHJP7PRYGKAKBZUGG2W/action/replication_record"}},"created_at":"2026-07-05T09:30:22.501917+00:00","updated_at":"2026-07-05T09:30:22.501917+00:00"}