{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:PTKH33VTRVIX3AEVT6IRMFPY7F","short_pith_number":"pith:PTKH33VT","schema_version":"1.0","canonical_sha256":"7cd47deeb38d517d80959f911615f8f95d8cb5d7afe0cdaaef4386fce28be04e","source":{"kind":"arxiv","id":"2407.03876","version":3},"attestation_state":"computed","paper":{"title":"Automated Progressive Red Teaming","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CR","authors_text":"Bojian Jiang, Deyi Xiong, Qing Yang, Tianhao Shen, Tong Wu, Yi Jing","submitted_at":"2024-07-04T12:14:27Z","abstract_excerpt":"Ensuring the safety of large language models (LLMs) is paramount, yet identifying potential vulnerabilities is challenging. While manual red teaming is effective, it is time-consuming, costly and lacks scalability. Automated red teaming (ART) offers a more cost-effective alternative, automatically generating adversarial prompts to expose LLM vulnerabilities. However, in current ART efforts, a robust framework is absent, which explicitly frames red teaming as an effectively learnable task. To address this gap, we propose Automated Progressive Red Teaming (APRT) as an effectively learnable frame"},"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":"2407.03876","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2024-07-04T12:14:27Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"67ed36fbc7f51dec52e4119730b238c6c0654814a434f81385562d481115e2f5","abstract_canon_sha256":"c2b1c8da39b2edc94dc6643498dbe488f15b96f2cc66a065ba6cb792ec156d92"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:52:38.530798Z","signature_b64":"eodOgCLzkZ6HxdzeKO4Y22OJfyaZFmpMn9+oyA2NLWR1J6n+/1J/XYmnOogbaa3vo0sd7nF1VgXksC4xQMYfAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7cd47deeb38d517d80959f911615f8f95d8cb5d7afe0cdaaef4386fce28be04e","last_reissued_at":"2026-07-05T09:52:38.530323Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:52:38.530323Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Automated Progressive Red Teaming","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CR","authors_text":"Bojian Jiang, Deyi Xiong, Qing Yang, Tianhao Shen, Tong Wu, Yi Jing","submitted_at":"2024-07-04T12:14:27Z","abstract_excerpt":"Ensuring the safety of large language models (LLMs) is paramount, yet identifying potential vulnerabilities is challenging. While manual red teaming is effective, it is time-consuming, costly and lacks scalability. Automated red teaming (ART) offers a more cost-effective alternative, automatically generating adversarial prompts to expose LLM vulnerabilities. However, in current ART efforts, a robust framework is absent, which explicitly frames red teaming as an effectively learnable task. To address this gap, we propose Automated Progressive Red Teaming (APRT) as an effectively learnable frame"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.03876","kind":"arxiv","version":3},"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/2407.03876/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":"2407.03876","created_at":"2026-07-05T09:52:38.530392+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.03876v3","created_at":"2026-07-05T09:52:38.530392+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.03876","created_at":"2026-07-05T09:52:38.530392+00:00"},{"alias_kind":"pith_short_12","alias_value":"PTKH33VTRVIX","created_at":"2026-07-05T09:52:38.530392+00:00"},{"alias_kind":"pith_short_16","alias_value":"PTKH33VTRVIX3AEV","created_at":"2026-07-05T09:52:38.530392+00:00"},{"alias_kind":"pith_short_8","alias_value":"PTKH33VT","created_at":"2026-07-05T09:52:38.530392+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.25476","citing_title":"A Red Teaming Framework for Large Language Models: A Case Study on Faithfulness Evaluation","ref_index":57,"is_internal_anchor":false},{"citing_arxiv_id":"2605.24949","citing_title":"APT-Agent: Automated Penetration Testing using Large Language Models","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2604.21159","citing_title":"Adaptive Instruction Composition for Automated LLM Red-Teaming","ref_index":47,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PTKH33VTRVIX3AEVT6IRMFPY7F","json":"https://pith.science/pith/PTKH33VTRVIX3AEVT6IRMFPY7F.json","graph_json":"https://pith.science/api/pith-number/PTKH33VTRVIX3AEVT6IRMFPY7F/graph.json","events_json":"https://pith.science/api/pith-number/PTKH33VTRVIX3AEVT6IRMFPY7F/events.json","paper":"https://pith.science/paper/PTKH33VT"},"agent_actions":{"view_html":"https://pith.science/pith/PTKH33VTRVIX3AEVT6IRMFPY7F","download_json":"https://pith.science/pith/PTKH33VTRVIX3AEVT6IRMFPY7F.json","view_paper":"https://pith.science/paper/PTKH33VT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.03876&json=true","fetch_graph":"https://pith.science/api/pith-number/PTKH33VTRVIX3AEVT6IRMFPY7F/graph.json","fetch_events":"https://pith.science/api/pith-number/PTKH33VTRVIX3AEVT6IRMFPY7F/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PTKH33VTRVIX3AEVT6IRMFPY7F/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PTKH33VTRVIX3AEVT6IRMFPY7F/action/storage_attestation","attest_author":"https://pith.science/pith/PTKH33VTRVIX3AEVT6IRMFPY7F/action/author_attestation","sign_citation":"https://pith.science/pith/PTKH33VTRVIX3AEVT6IRMFPY7F/action/citation_signature","submit_replication":"https://pith.science/pith/PTKH33VTRVIX3AEVT6IRMFPY7F/action/replication_record"}},"created_at":"2026-07-05T09:52:38.530392+00:00","updated_at":"2026-07-05T09:52:38.530392+00:00"}