{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:GRDN7ZXEW4GV47JK7TGIKYDALR","short_pith_number":"pith:GRDN7ZXE","schema_version":"1.0","canonical_sha256":"3446dfe6e4b70d5e7d2afccc8560605c75ffa5d7493ccca4c3a76b6125c8770f","source":{"kind":"arxiv","id":"2408.14866","version":2},"attestation_state":"computed","paper":{"title":"Advancing Adversarial Suffix Transfer Learning on Aligned Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CR","cs.LG"],"primary_cat":"cs.CL","authors_text":"Hongfu Liu, Michael Shieh, Ye Wang, Yuxi Xie","submitted_at":"2024-08-27T08:38:48Z","abstract_excerpt":"Language Language Models (LLMs) face safety concerns due to potential misuse by malicious users. Recent red-teaming efforts have identified adversarial suffixes capable of jailbreaking LLMs using the gradient-based search algorithm Greedy Coordinate Gradient (GCG). However, GCG struggles with computational inefficiency, limiting further investigations regarding suffix transferability and scalability across models and data. In this work, we bridge the connection between search efficiency and suffix transferability. We propose a two-stage transfer learning framework, DeGCG, which decouples the s"},"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":"2408.14866","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-08-27T08:38:48Z","cross_cats_sorted":["cs.CR","cs.LG"],"title_canon_sha256":"b9b6227b38f8897b0bc85c639ec0c5dc86744647887405fe4a4f805c40b5c44c","abstract_canon_sha256":"1c8a2e3cfa0bd4251d0817100d9a1ccfc5ac2558016e062f07aa8e406ca67a21"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:16:20.699787Z","signature_b64":"DA0Cx3syhfNFVrMBzhfEewnpP/k9wy3YtKpqKNyrvi+Ch0p4lTUk3xaC4BrRqUY00C0myXpaNr2txlsfGOsIBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3446dfe6e4b70d5e7d2afccc8560605c75ffa5d7493ccca4c3a76b6125c8770f","last_reissued_at":"2026-07-05T09:16:20.699357Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:16:20.699357Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Advancing Adversarial Suffix Transfer Learning on Aligned Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CR","cs.LG"],"primary_cat":"cs.CL","authors_text":"Hongfu Liu, Michael Shieh, Ye Wang, Yuxi Xie","submitted_at":"2024-08-27T08:38:48Z","abstract_excerpt":"Language Language Models (LLMs) face safety concerns due to potential misuse by malicious users. Recent red-teaming efforts have identified adversarial suffixes capable of jailbreaking LLMs using the gradient-based search algorithm Greedy Coordinate Gradient (GCG). However, GCG struggles with computational inefficiency, limiting further investigations regarding suffix transferability and scalability across models and data. In this work, we bridge the connection between search efficiency and suffix transferability. We propose a two-stage transfer learning framework, DeGCG, which decouples the s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.14866","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/2408.14866/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":"2408.14866","created_at":"2026-07-05T09:16:20.699422+00:00"},{"alias_kind":"arxiv_version","alias_value":"2408.14866v2","created_at":"2026-07-05T09:16:20.699422+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.14866","created_at":"2026-07-05T09:16:20.699422+00:00"},{"alias_kind":"pith_short_12","alias_value":"GRDN7ZXEW4GV","created_at":"2026-07-05T09:16:20.699422+00:00"},{"alias_kind":"pith_short_16","alias_value":"GRDN7ZXEW4GV47JK","created_at":"2026-07-05T09:16:20.699422+00:00"},{"alias_kind":"pith_short_8","alias_value":"GRDN7ZXE","created_at":"2026-07-05T09:16:20.699422+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/GRDN7ZXEW4GV47JK7TGIKYDALR","json":"https://pith.science/pith/GRDN7ZXEW4GV47JK7TGIKYDALR.json","graph_json":"https://pith.science/api/pith-number/GRDN7ZXEW4GV47JK7TGIKYDALR/graph.json","events_json":"https://pith.science/api/pith-number/GRDN7ZXEW4GV47JK7TGIKYDALR/events.json","paper":"https://pith.science/paper/GRDN7ZXE"},"agent_actions":{"view_html":"https://pith.science/pith/GRDN7ZXEW4GV47JK7TGIKYDALR","download_json":"https://pith.science/pith/GRDN7ZXEW4GV47JK7TGIKYDALR.json","view_paper":"https://pith.science/paper/GRDN7ZXE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2408.14866&json=true","fetch_graph":"https://pith.science/api/pith-number/GRDN7ZXEW4GV47JK7TGIKYDALR/graph.json","fetch_events":"https://pith.science/api/pith-number/GRDN7ZXEW4GV47JK7TGIKYDALR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GRDN7ZXEW4GV47JK7TGIKYDALR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GRDN7ZXEW4GV47JK7TGIKYDALR/action/storage_attestation","attest_author":"https://pith.science/pith/GRDN7ZXEW4GV47JK7TGIKYDALR/action/author_attestation","sign_citation":"https://pith.science/pith/GRDN7ZXEW4GV47JK7TGIKYDALR/action/citation_signature","submit_replication":"https://pith.science/pith/GRDN7ZXEW4GV47JK7TGIKYDALR/action/replication_record"}},"created_at":"2026-07-05T09:16:20.699422+00:00","updated_at":"2026-07-05T09:16:20.699422+00:00"}