{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:VSW6OGZPLEEPTZKOOAFL2IM36N","short_pith_number":"pith:VSW6OGZP","schema_version":"1.0","canonical_sha256":"acade71b2f5908f9e54e700abd219bf34b1d1b311ef2d5ab4a0b8e7f3e2f14e6","source":{"kind":"arxiv","id":"2606.21082","version":1},"attestation_state":"computed","paper":{"title":"Scalable Hierarchical Attention Transformers for Multi-Turn Jailbreak Detection in Long Conversations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CR"],"primary_cat":"cs.CL","authors_text":"Chenhui Hu, Muhammed Salih, Subramanian Srinivasan, Sudipto Guha","submitted_at":"2026-06-19T04:05:43Z","abstract_excerpt":"Multi-turn jailbreaks can evade turn-level moderation by spreading unsafe intent across a dialogue through gradual escalation, reframing, and role manipulation. We address multi-turn jailbreak detection as a conversation-level classification problem and introduce an efficient hierarchical detector that avoids expensive long-context concatenation while retaining cross-turn reasoning. The model encodes individual turns to form compact turn representations and applies a lightweight conversation module that captures dialogue dynamics and selectively attends to fine-grained evidence when needed. On"},"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":"2606.21082","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-19T04:05:43Z","cross_cats_sorted":["cs.AI","cs.CR"],"title_canon_sha256":"ec52a50812c39eae5c5bd1dc98f3bd9851d4b1c28235c73272cac02911faa691","abstract_canon_sha256":"81adda67618d78b5a141ef0b193bfb598f8d0198fb90332da4f90a25eeca15f9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T01:12:29.235724Z","signature_b64":"C+5wFsPm9nanIp8Fw/6cZByq/vOhPJvFgRGN6Rl0xuUKPC4avpxC9Q6gsG0RW2Zh70EgtkpmbwzEKBEis/MRBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"acade71b2f5908f9e54e700abd219bf34b1d1b311ef2d5ab4a0b8e7f3e2f14e6","last_reissued_at":"2026-06-23T01:12:29.235234Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T01:12:29.235234Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Scalable Hierarchical Attention Transformers for Multi-Turn Jailbreak Detection in Long Conversations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CR"],"primary_cat":"cs.CL","authors_text":"Chenhui Hu, Muhammed Salih, Subramanian Srinivasan, Sudipto Guha","submitted_at":"2026-06-19T04:05:43Z","abstract_excerpt":"Multi-turn jailbreaks can evade turn-level moderation by spreading unsafe intent across a dialogue through gradual escalation, reframing, and role manipulation. We address multi-turn jailbreak detection as a conversation-level classification problem and introduce an efficient hierarchical detector that avoids expensive long-context concatenation while retaining cross-turn reasoning. The model encodes individual turns to form compact turn representations and applies a lightweight conversation module that captures dialogue dynamics and selectively attends to fine-grained evidence when needed. On"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.21082","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/2606.21082/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":"2606.21082","created_at":"2026-06-23T01:12:29.235323+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.21082v1","created_at":"2026-06-23T01:12:29.235323+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.21082","created_at":"2026-06-23T01:12:29.235323+00:00"},{"alias_kind":"pith_short_12","alias_value":"VSW6OGZPLEEP","created_at":"2026-06-23T01:12:29.235323+00:00"},{"alias_kind":"pith_short_16","alias_value":"VSW6OGZPLEEPTZKO","created_at":"2026-06-23T01:12:29.235323+00:00"},{"alias_kind":"pith_short_8","alias_value":"VSW6OGZP","created_at":"2026-06-23T01:12:29.235323+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/VSW6OGZPLEEPTZKOOAFL2IM36N","json":"https://pith.science/pith/VSW6OGZPLEEPTZKOOAFL2IM36N.json","graph_json":"https://pith.science/api/pith-number/VSW6OGZPLEEPTZKOOAFL2IM36N/graph.json","events_json":"https://pith.science/api/pith-number/VSW6OGZPLEEPTZKOOAFL2IM36N/events.json","paper":"https://pith.science/paper/VSW6OGZP"},"agent_actions":{"view_html":"https://pith.science/pith/VSW6OGZPLEEPTZKOOAFL2IM36N","download_json":"https://pith.science/pith/VSW6OGZPLEEPTZKOOAFL2IM36N.json","view_paper":"https://pith.science/paper/VSW6OGZP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.21082&json=true","fetch_graph":"https://pith.science/api/pith-number/VSW6OGZPLEEPTZKOOAFL2IM36N/graph.json","fetch_events":"https://pith.science/api/pith-number/VSW6OGZPLEEPTZKOOAFL2IM36N/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VSW6OGZPLEEPTZKOOAFL2IM36N/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VSW6OGZPLEEPTZKOOAFL2IM36N/action/storage_attestation","attest_author":"https://pith.science/pith/VSW6OGZPLEEPTZKOOAFL2IM36N/action/author_attestation","sign_citation":"https://pith.science/pith/VSW6OGZPLEEPTZKOOAFL2IM36N/action/citation_signature","submit_replication":"https://pith.science/pith/VSW6OGZPLEEPTZKOOAFL2IM36N/action/replication_record"}},"created_at":"2026-06-23T01:12:29.235323+00:00","updated_at":"2026-06-23T01:12:29.235323+00:00"}