{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:2M623PTWVQHC5FSKJDCC5N3I6A","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":"9073112c5e28e4eca530635ba7f84738b8a1430a3b7ecf8fcfdb1d3692c4794e","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-24T05:49:14Z","title_canon_sha256":"15ccc32d25cb3361093224b610e1957551be2829b49ff9811bc2fbd07859d725"},"schema_version":"1.0","source":{"id":"2607.20472","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.20472","created_at":"2026-07-24T00:23:17Z"},{"alias_kind":"arxiv_version","alias_value":"2607.20472v1","created_at":"2026-07-24T00:23:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.20472","created_at":"2026-07-24T00:23:17Z"},{"alias_kind":"pith_short_12","alias_value":"2M623PTWVQHC","created_at":"2026-07-24T00:23:17Z"},{"alias_kind":"pith_short_16","alias_value":"2M623PTWVQHC5FSK","created_at":"2026-07-24T00:23:17Z"},{"alias_kind":"pith_short_8","alias_value":"2M623PTW","created_at":"2026-07-24T00:23:17Z"}],"graph_snapshots":[{"event_id":"sha256:cc0be1e9f409e560cf2fa01546e9ec946fa121e3b4ae4b6a7979035aa0076772","target":"graph","created_at":"2026-07-24T00:23:17Z","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/2607.20472/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"When a user asks a language model something harmful, is it a genuine attack or a misunderstood but well-meaning question? This ambiguity is one of the central challenges of LLM safety. A model that assumes the worst harms legitimate users; one that assumes the best is easily exploited. The problem is compounded in multi-turn dialogue, where an attacker's true intent may only reveal itself gradually across many exchanges, yet existing safety frameworks apply a contextual bandit treatment, ignoring the trajectory of the conversation.\n  To that end, we propose Dialogue Critic Guided Sampling (DCG","authors_text":"Arunesh Sinha, Pradeep Varakantham, Roman Belaire","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-24T05:49:14Z","title":"Robust Critics: Defending LLMs Against Multi-Turn Attacks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.20472","kind":"arxiv","version":1},"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:e48e53b7ef8354a1fad7c471c204dbc36249e20ae6367a82ab96d86703245af2","target":"record","created_at":"2026-07-24T00:23:17Z","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":"9073112c5e28e4eca530635ba7f84738b8a1430a3b7ecf8fcfdb1d3692c4794e","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-24T05:49:14Z","title_canon_sha256":"15ccc32d25cb3361093224b610e1957551be2829b49ff9811bc2fbd07859d725"},"schema_version":"1.0","source":{"id":"2607.20472","kind":"arxiv","version":1}},"canonical_sha256":"d33dadbe76ac0e2e964a48c42eb768f007038ad43377143b27d1ee3ad7f17084","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d33dadbe76ac0e2e964a48c42eb768f007038ad43377143b27d1ee3ad7f17084","first_computed_at":"2026-07-24T00:23:17.572244Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-24T00:23:17.572244Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"z3nRdvi2Nm+NPVRYWsX45q+V1r+DK5yLJmLTGLucnoVGOHYOU2AmJvKbnHIgHZx/x6zAlEG5R0IJSrF0QlcXBg==","signature_status":"signed_v1","signed_at":"2026-07-24T00:23:17.573107Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.20472","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e48e53b7ef8354a1fad7c471c204dbc36249e20ae6367a82ab96d86703245af2","sha256:cc0be1e9f409e560cf2fa01546e9ec946fa121e3b4ae4b6a7979035aa0076772"],"state_sha256":"53650c296543dc4e1ede2d3dd1a617584ee72fe811bca71af7df3c704d919768"}