{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:AYI57RKG6PLI6FYD2DDFFVRYQ2","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":"bd69cfd2004eb5d7eadca2b180c06e8a4e37440c87e62e0358357439c57ad83a","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-04T02:36:41Z","title_canon_sha256":"68c9cceef9d821975e3a1f951bc1884e1e15dee9384f3fcd1f50e0463799e660"},"schema_version":"1.0","source":{"id":"2606.05614","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.05614","created_at":"2026-06-05T01:14:56Z"},{"alias_kind":"arxiv_version","alias_value":"2606.05614v1","created_at":"2026-06-05T01:14:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.05614","created_at":"2026-06-05T01:14:56Z"},{"alias_kind":"pith_short_12","alias_value":"AYI57RKG6PLI","created_at":"2026-06-05T01:14:56Z"},{"alias_kind":"pith_short_16","alias_value":"AYI57RKG6PLI6FYD","created_at":"2026-06-05T01:14:56Z"},{"alias_kind":"pith_short_8","alias_value":"AYI57RKG","created_at":"2026-06-05T01:14:56Z"}],"graph_snapshots":[{"event_id":"sha256:69daa6c01cee0c2a4778a73255f02aa683186464cffebbb086ef07c03b207593","target":"graph","created_at":"2026-06-05T01:14:56Z","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/2606.05614/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) are rigorously aligned to refuse harmful requests, a process that inherently cultivates a latent capacity to evaluate and recognize unsafe content. In this work, we reveal that this advanced safety awareness inadvertently introduces a fatal vulnerability. We introduce Posterior Attack, a single-query jailbreak that bypasses guardrails by prompting the model to generate the exact harmful response its internal classifier would normally flag as unsafe. Through extensive empirical evaluation across 30 open-source LLMs (up to 35B parameters in size) and frontier models ","authors_text":"Hai V. Le, Long P. Hoang, Shaoyang Xu, Wei Lu, Wenxuan Zhang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-04T02:36:41Z","title":"Safety Paradox: How Enhanced Safety Awareness Leaves LLMs Vulnerable to Posterior Attack"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.05614","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":"sha256:67b74db5062c370d79496718de779385385a1ffb3865358ae4021aa15cfb81ff","target":"record","created_at":"2026-06-05T01:14:56Z","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":"bd69cfd2004eb5d7eadca2b180c06e8a4e37440c87e62e0358357439c57ad83a","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-04T02:36:41Z","title_canon_sha256":"68c9cceef9d821975e3a1f951bc1884e1e15dee9384f3fcd1f50e0463799e660"},"schema_version":"1.0","source":{"id":"2606.05614","kind":"arxiv","version":1}},"canonical_sha256":"0611dfc546f3d68f1703d0c652d638869c032ca1927b9a326516086017499e1b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0611dfc546f3d68f1703d0c652d638869c032ca1927b9a326516086017499e1b","first_computed_at":"2026-06-05T01:14:56.484785Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-05T01:14:56.484785Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"E/e0QGR7kzb5+I+ZZX/c5zYs8H9uJQ4/BFGXRxZf4OJODMPTwIrgRI+avVkFbL4fPXAMRR9VXyO2QJF+Kr+ABg==","signature_status":"signed_v1","signed_at":"2026-06-05T01:14:56.485232Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.05614","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:67b74db5062c370d79496718de779385385a1ffb3865358ae4021aa15cfb81ff","sha256:69daa6c01cee0c2a4778a73255f02aa683186464cffebbb086ef07c03b207593"],"state_sha256":"122f49c32ed16e78f860d3820249dcd62f48657118fd9b0127cf810380c14ea7"}