{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:ZXP2FNIOQZ37DIJQVRKXQYW7IQ","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"c1db50f0db0d5160021c592ceae183687aa75ed162f924532902235e82519ec6","cross_cats_sorted":["cs.AI","cs.CL","cs.LG","cs.MA"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-04-15T16:11:28Z","title_canon_sha256":"e4f2f540fb2381a4ba8fb960721fa248a03f463c7dcde5aac1ff1daad5dea625"},"schema_version":"1.0","source":{"id":"2504.13203","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.13203","created_at":"2026-07-05T11:57:56Z"},{"alias_kind":"arxiv_version","alias_value":"2504.13203v2","created_at":"2026-07-05T11:57:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.13203","created_at":"2026-07-05T11:57:56Z"},{"alias_kind":"pith_short_12","alias_value":"ZXP2FNIOQZ37","created_at":"2026-07-05T11:57:56Z"},{"alias_kind":"pith_short_16","alias_value":"ZXP2FNIOQZ37DIJQ","created_at":"2026-07-05T11:57:56Z"},{"alias_kind":"pith_short_8","alias_value":"ZXP2FNIO","created_at":"2026-07-05T11:57:56Z"}],"graph_snapshots":[{"event_id":"sha256:075dda73f3b84ff8996656af3fd52a4d16d2d64a92dfc9cd55994f8726e47588","target":"graph","created_at":"2026-07-05T11:57:56Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2504.13203/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Multi-turn interactions with language models (LMs) pose critical safety risks, as harmful intent can be strategically spread across exchanges. Yet, the vast majority of prior work has focused on single-turn safety, while adaptability and diversity remain among the key challenges of multi-turn red-teaming. To address these challenges, we present X-Teaming, a scalable framework that systematically explores how seemingly harmless interactions escalate into harmful outcomes and generates corresponding attack scenarios. X-Teaming employs collaborative agents for planning, attack optimization, and v","authors_text":"Genglin Liu, Hamid Palangi, James Shiffer, Kai-Wei Chang, Liwei Jiang, Md Rizwan Parvez, Saadia Gabriel, Salman Rahman, Sheriff Issaka, Yejin Choi","cross_cats":["cs.AI","cs.CL","cs.LG","cs.MA"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-04-15T16:11:28Z","title":"X-Teaming: Multi-Turn Jailbreaks and Defenses with Adaptive Multi-Agents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.13203","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:e24cae874c81cfa215afc6728de865db707353bd9b1d719524c8676710f90405","target":"record","created_at":"2026-07-05T11:57:56Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"c1db50f0db0d5160021c592ceae183687aa75ed162f924532902235e82519ec6","cross_cats_sorted":["cs.AI","cs.CL","cs.LG","cs.MA"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-04-15T16:11:28Z","title_canon_sha256":"e4f2f540fb2381a4ba8fb960721fa248a03f463c7dcde5aac1ff1daad5dea625"},"schema_version":"1.0","source":{"id":"2504.13203","kind":"arxiv","version":2}},"canonical_sha256":"cddfa2b50e8677f1a130ac557862df443f56e3791cf0f9d7bd769e9ebc119d42","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"cddfa2b50e8677f1a130ac557862df443f56e3791cf0f9d7bd769e9ebc119d42","first_computed_at":"2026-07-05T11:57:56.053367Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:57:56.053367Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"4cQZUbO+pnvc8kskp5eD++cqgNhgn0TLE+B+Egit58XKeISyH34UEGSaiFsSd5CcuKg+BEvXyuJAGCbsxq5QAA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:57:56.053839Z","signed_message":"canonical_sha256_bytes"},"source_id":"2504.13203","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e24cae874c81cfa215afc6728de865db707353bd9b1d719524c8676710f90405","sha256:075dda73f3b84ff8996656af3fd52a4d16d2d64a92dfc9cd55994f8726e47588"],"state_sha256":"ecd5f0f6e36073ac496c09cbeb41fddaa9ffe0342bc7555da3fda9b5ff137faf"}