{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:Z27HHRVP24HL3MBEVWI367U7CR","short_pith_number":"pith:Z27HHRVP","schema_version":"1.0","canonical_sha256":"cebe73c6afd70ebdb024ad91bf7e9f14613bafb92037e90c5342cd0289f1df21","source":{"kind":"arxiv","id":"2607.11151","version":1},"attestation_state":"computed","paper":{"title":"AMT-X: Phase-Structured Multi-Turn Red-Teaming with Checklist-Gated Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Alex Leung, Kentaroh Toyoda, Yi Ting Shen","submitted_at":"2026-07-13T06:46:56Z","abstract_excerpt":"Safety evaluation of large language models (LLMs) relies largely on single-turn attack datasets and single-judge scoring, underestimating risk from adaptive multi-turn adversaries and reporting a single success rate that does not separate partially actionable outputs from those carrying complete operational detail. We propose AMT-X (Adaptive Multi-Turn Exploitation), a phase-structured multi-turn red-teaming framework. Unlike prior multi-turn attacks that rely on ad hoc escalation or free-form per-goal plans, AMT-X casts the attack as an explicit, reproducible multi-phase state machine driven "},"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":"2607.11151","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2026-07-13T06:46:56Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ec9743e616178085fe8b47bc0a660f9319a981c258dbfa206dd569e7342de9ac","abstract_canon_sha256":"3ed133a5b7c98e9acbae4519d79bd36ad605a262e1b9031ab574c872f4d19237"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-14T01:22:19.202202Z","signature_b64":"46fEnbgT8uD5eeuKNE01HG6oX7jW0Y2XIQZQIDaBnfbe2+4PWnDMiQ+iEHlCYSz2wJVQ67jkOSgelrpOUzH0CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cebe73c6afd70ebdb024ad91bf7e9f14613bafb92037e90c5342cd0289f1df21","last_reissued_at":"2026-07-14T01:22:19.201260Z","signature_status":"signed_v1","first_computed_at":"2026-07-14T01:22:19.201260Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"AMT-X: Phase-Structured Multi-Turn Red-Teaming with Checklist-Gated Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Alex Leung, Kentaroh Toyoda, Yi Ting Shen","submitted_at":"2026-07-13T06:46:56Z","abstract_excerpt":"Safety evaluation of large language models (LLMs) relies largely on single-turn attack datasets and single-judge scoring, underestimating risk from adaptive multi-turn adversaries and reporting a single success rate that does not separate partially actionable outputs from those carrying complete operational detail. We propose AMT-X (Adaptive Multi-Turn Exploitation), a phase-structured multi-turn red-teaming framework. Unlike prior multi-turn attacks that rely on ad hoc escalation or free-form per-goal plans, AMT-X casts the attack as an explicit, reproducible multi-phase state machine driven "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.11151","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/2607.11151/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":"2607.11151","created_at":"2026-07-14T01:22:19.201749+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.11151v1","created_at":"2026-07-14T01:22:19.201749+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.11151","created_at":"2026-07-14T01:22:19.201749+00:00"},{"alias_kind":"pith_short_12","alias_value":"Z27HHRVP24HL","created_at":"2026-07-14T01:22:19.201749+00:00"},{"alias_kind":"pith_short_16","alias_value":"Z27HHRVP24HL3MBE","created_at":"2026-07-14T01:22:19.201749+00:00"},{"alias_kind":"pith_short_8","alias_value":"Z27HHRVP","created_at":"2026-07-14T01:22:19.201749+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/Z27HHRVP24HL3MBEVWI367U7CR","json":"https://pith.science/pith/Z27HHRVP24HL3MBEVWI367U7CR.json","graph_json":"https://pith.science/api/pith-number/Z27HHRVP24HL3MBEVWI367U7CR/graph.json","events_json":"https://pith.science/api/pith-number/Z27HHRVP24HL3MBEVWI367U7CR/events.json","paper":"https://pith.science/paper/Z27HHRVP"},"agent_actions":{"view_html":"https://pith.science/pith/Z27HHRVP24HL3MBEVWI367U7CR","download_json":"https://pith.science/pith/Z27HHRVP24HL3MBEVWI367U7CR.json","view_paper":"https://pith.science/paper/Z27HHRVP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.11151&json=true","fetch_graph":"https://pith.science/api/pith-number/Z27HHRVP24HL3MBEVWI367U7CR/graph.json","fetch_events":"https://pith.science/api/pith-number/Z27HHRVP24HL3MBEVWI367U7CR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Z27HHRVP24HL3MBEVWI367U7CR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Z27HHRVP24HL3MBEVWI367U7CR/action/storage_attestation","attest_author":"https://pith.science/pith/Z27HHRVP24HL3MBEVWI367U7CR/action/author_attestation","sign_citation":"https://pith.science/pith/Z27HHRVP24HL3MBEVWI367U7CR/action/citation_signature","submit_replication":"https://pith.science/pith/Z27HHRVP24HL3MBEVWI367U7CR/action/replication_record"}},"created_at":"2026-07-14T01:22:19.201749+00:00","updated_at":"2026-07-14T01:22:19.201749+00:00"}