{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:ZJX7T3SCKCLFJECWBFU7R3V6AU","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":"b7bb907ea771cea181b529dfd0c99b325996ed9c4bcbe237b15e31127acfb735","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-11T16:38:01Z","title_canon_sha256":"9530864ec63dd03518b2d912e2615b375202f2bf9b02aaaf0023c5bbc4dd96e9"},"schema_version":"1.0","source":{"id":"2410.08968","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.08968","created_at":"2026-07-05T10:23:23Z"},{"alias_kind":"arxiv_version","alias_value":"2410.08968v2","created_at":"2026-07-05T10:23:23Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.08968","created_at":"2026-07-05T10:23:23Z"},{"alias_kind":"pith_short_12","alias_value":"ZJX7T3SCKCLF","created_at":"2026-07-05T10:23:23Z"},{"alias_kind":"pith_short_16","alias_value":"ZJX7T3SCKCLFJECW","created_at":"2026-07-05T10:23:23Z"},{"alias_kind":"pith_short_8","alias_value":"ZJX7T3SC","created_at":"2026-07-05T10:23:23Z"}],"graph_snapshots":[{"event_id":"sha256:cbc8b1593e0c0c85d62596d7283bd7ad6f547bad8fe708e172c3bae71ca6ac4d","target":"graph","created_at":"2026-07-05T10:23:23Z","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/2410.08968/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The current paradigm for safety alignment of large language models (LLMs) follows a one-size-fits-all approach: the model refuses to interact with any content deemed unsafe by the model provider. This approach lacks flexibility in the face of varying social norms across cultures and regions. In addition, users may have diverse safety needs, making a model with static safety standards too restrictive to be useful, as well as too costly to be re-aligned.\n  We propose Controllable Safety Alignment (CoSA), a framework designed to adapt models to diverse safety requirements without re-training. Ins","authors_text":"Ahmed Elgohary, Ahmed Magooda, Benjamin Van Durme, Daniel Khashabi, Jingyu Zhang","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-11T16:38:01Z","title":"Controllable Safety Alignment: Inference-Time Adaptation to Diverse Safety Requirements"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.08968","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":"sha256:7441b35a3478b3c730af49a9bcc3a5d19f684fef931e438c3c5b1b1c8fc611bc","target":"record","created_at":"2026-07-05T10:23:23Z","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":"b7bb907ea771cea181b529dfd0c99b325996ed9c4bcbe237b15e31127acfb735","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-11T16:38:01Z","title_canon_sha256":"9530864ec63dd03518b2d912e2615b375202f2bf9b02aaaf0023c5bbc4dd96e9"},"schema_version":"1.0","source":{"id":"2410.08968","kind":"arxiv","version":2}},"canonical_sha256":"ca6ff9ee4250965490560969f8eebe0518490b28a27b52a11d1e9aa3fa4a56dc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ca6ff9ee4250965490560969f8eebe0518490b28a27b52a11d1e9aa3fa4a56dc","first_computed_at":"2026-07-05T10:23:23.978439Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:23:23.978439Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"CMB5ZXrc9e0Btm8LjNKJY6Ytjj4dI6fX5SQjrfMqnoqT1NMeZxJY3BMAJza1Pz2AH9XjGYWJzDE8kNl+IJC2Cw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:23:23.978965Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.08968","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7441b35a3478b3c730af49a9bcc3a5d19f684fef931e438c3c5b1b1c8fc611bc","sha256:cbc8b1593e0c0c85d62596d7283bd7ad6f547bad8fe708e172c3bae71ca6ac4d"],"state_sha256":"63c18c4b26d480832e1d06bb290b14f1ab2b68c190b4964b74e8836b378587ad"}