{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:4MTKL7CL53P4OA7STXULTIO6KV","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":"d208a78f745035370726c25082dc6401957e07d7c203b998536243dc5db75642","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-05-22T16:51:07Z","title_canon_sha256":"b5ea66ea0da5e85128fe9172ea69ec7635fddf8980addcfd1e7767deef128b00"},"schema_version":"1.0","source":{"id":"2405.13820","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.13820","created_at":"2026-07-05T09:50:02Z"},{"alias_kind":"arxiv_version","alias_value":"2405.13820v2","created_at":"2026-07-05T09:50:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.13820","created_at":"2026-07-05T09:50:02Z"},{"alias_kind":"pith_short_12","alias_value":"4MTKL7CL53P4","created_at":"2026-07-05T09:50:02Z"},{"alias_kind":"pith_short_16","alias_value":"4MTKL7CL53P4OA7S","created_at":"2026-07-05T09:50:02Z"},{"alias_kind":"pith_short_8","alias_value":"4MTKL7CL","created_at":"2026-07-05T09:50:02Z"}],"graph_snapshots":[{"event_id":"sha256:faa1cdc73c7323b8322950d820cd60ac5a162d896c5ea3ee9407b7055becdaf0","target":"graph","created_at":"2026-07-05T09:50:02Z","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/2405.13820/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Safety alignment of large language models (LLMs) has been gaining increasing attention. However, current safety-aligned LLMs suffer from the fragile and imbalanced safety mechanisms, which can still be induced to generate unsafe responses, exhibit over-safety by rejecting safe user inputs, and fail to preserve general utility after safety alignment. To this end, we propose a novel post safety alignment (PSA) method to address these inherent and emerging safety challenges, including safety enhancement, over-safety mitigation, and utility preservation. In specific, we introduce \\textsc{SafePatch","authors_text":"Bing Qin, Jiahe Guo, Tat-Seng Chua, Ting Liu, Weixiang Zhao, Xingyu Sui, Yang Deng, Yanyan Zhao, Yulin Hu, Zhuojun Li","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-05-22T16:51:07Z","title":"Towards Comprehensive Post Safety Alignment of Large Language Models via Safety Patching"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.13820","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:f571a21715768a81a43598a4f0748e14eb1bcce1d1c4e5a6d7b02ace0156f062","target":"record","created_at":"2026-07-05T09:50:02Z","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":"d208a78f745035370726c25082dc6401957e07d7c203b998536243dc5db75642","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-05-22T16:51:07Z","title_canon_sha256":"b5ea66ea0da5e85128fe9172ea69ec7635fddf8980addcfd1e7767deef128b00"},"schema_version":"1.0","source":{"id":"2405.13820","kind":"arxiv","version":2}},"canonical_sha256":"e326a5fc4beedfc703f29de8b9a1de555a2c83ec45a2fa8c8d603a778db12dae","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e326a5fc4beedfc703f29de8b9a1de555a2c83ec45a2fa8c8d603a778db12dae","first_computed_at":"2026-07-05T09:50:02.587607Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:50:02.587607Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"625dBwaeLXYTmIk1fLHOAGh5HjwLT6uN3i9KbeyuUMOeNsIzTX2iDiUSdI486zQ4U+81CrCeumB9ZnJ2ZRhWBg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:50:02.588092Z","signed_message":"canonical_sha256_bytes"},"source_id":"2405.13820","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f571a21715768a81a43598a4f0748e14eb1bcce1d1c4e5a6d7b02ace0156f062","sha256:faa1cdc73c7323b8322950d820cd60ac5a162d896c5ea3ee9407b7055becdaf0"],"state_sha256":"552162f6d57d9e11d4579ad6e7c075c7517eb44b4eac5ce125d3d5a708e2537b"}