{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:2SCG5PZG27A47AZGRBCKZPQ2TR","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":"4d0b3c79290c1ea72d9dcc58a8220f8dd6c2fc59648742d6eabf8cf417c0a7e5","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CY","submitted_at":"2024-10-10T15:06:12Z","title_canon_sha256":"b18522d17287bb9d119e41b431a40aba808ad975d157182807988ca42b89f69c"},"schema_version":"1.0","source":{"id":"2410.18991","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.18991","created_at":"2026-07-05T09:30:23Z"},{"alias_kind":"arxiv_version","alias_value":"2410.18991v2","created_at":"2026-07-05T09:30:23Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.18991","created_at":"2026-07-05T09:30:23Z"},{"alias_kind":"pith_short_12","alias_value":"2SCG5PZG27A4","created_at":"2026-07-05T09:30:23Z"},{"alias_kind":"pith_short_16","alias_value":"2SCG5PZG27A47AZG","created_at":"2026-07-05T09:30:23Z"},{"alias_kind":"pith_short_8","alias_value":"2SCG5PZG","created_at":"2026-07-05T09:30:23Z"}],"graph_snapshots":[{"event_id":"sha256:1a961265de242e714ed086339e0a4ec47b22aca156d4a040690cc103c3751d11","target":"graph","created_at":"2026-07-05T09:30:23Z","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/2410.18991/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We present the TRIAGE Benchmark, a novel machine ethics (ME) benchmark that tests LLMs' ability to make ethical decisions during mass casualty incidents. It uses real-world ethical dilemmas with clear solutions designed by medical professionals, offering a more realistic alternative to annotation-based benchmarks. TRIAGE incorporates various prompting styles to evaluate model performance across different contexts. Most models consistently outperformed random guessing, suggesting LLMs may support decision-making in triage scenarios. Neutral or factual scenario formulations led to the best perfo","authors_text":"Konstantin Hebenstreit, Matthias Samwald, Nathalie Maria Kirch","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CY","submitted_at":"2024-10-10T15:06:12Z","title":"TRIAGE: Ethical Benchmarking of AI Models Through Mass Casualty Simulations"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.18991","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":"sha256:6066d6cfc3594bb8b83dd484fbea9b9d0f150e7ac731fc343ffa50ff1628cd57","target":"record","created_at":"2026-07-05T09:30:23Z","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":"4d0b3c79290c1ea72d9dcc58a8220f8dd6c2fc59648742d6eabf8cf417c0a7e5","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CY","submitted_at":"2024-10-10T15:06:12Z","title_canon_sha256":"b18522d17287bb9d119e41b431a40aba808ad975d157182807988ca42b89f69c"},"schema_version":"1.0","source":{"id":"2410.18991","kind":"arxiv","version":2}},"canonical_sha256":"d4846ebf26d7c1cf83268844acbe1a9c482ec21d691d2d8b07aefd37de33f0b9","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d4846ebf26d7c1cf83268844acbe1a9c482ec21d691d2d8b07aefd37de33f0b9","first_computed_at":"2026-07-05T09:30:23.028460Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:30:23.028460Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"bI1yysqrQcyeyo0nDwa3zgXL9nZEqAvv4Qqt4+1UxCO0V2Jef8RwHx3iupLMxRcCXbPID9ULYt7O2ltEZm/qCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:30:23.028925Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.18991","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6066d6cfc3594bb8b83dd484fbea9b9d0f150e7ac731fc343ffa50ff1628cd57","sha256:1a961265de242e714ed086339e0a4ec47b22aca156d4a040690cc103c3751d11"],"state_sha256":"994a402b4beea4fb8ac330ed36263929991b06856b88440b3d539a456a310f45"}