{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:TX4NYGQLM2HTWDZD2YT4KZU3WH","short_pith_number":"pith:TX4NYGQL","schema_version":"1.0","canonical_sha256":"9df8dc1a0b668f3b0f23d627c5669bb1c12cc44d4cdde43e6f959737cd926fee","source":{"kind":"arxiv","id":"2505.21828","version":1},"attestation_state":"computed","paper":{"title":"SAGE-Eval: Evaluating LLMs for Systematic Generalizations of Safety Facts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Brenden M. Lake, Chen Yueh-Han, Guy Davidson","submitted_at":"2025-05-27T23:29:32Z","abstract_excerpt":"Do LLMs robustly generalize critical safety facts to novel situations? Lacking this ability is dangerous when users ask naive questions. For instance, \"I'm considering packing melon balls for my 10-month-old's lunch. What other foods would be good to include?\" Before offering food options, the LLM should warn that melon balls pose a choking hazard to toddlers, as documented by the CDC. Failing to provide such warnings could result in serious injuries or even death. To evaluate this, we introduce SAGE-Eval, SAfety-fact systematic GEneralization evaluation, the first benchmark that tests whether"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2505.21828","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-27T23:29:32Z","cross_cats_sorted":[],"title_canon_sha256":"472f34aa6e2f5795680aa5785c0258b35a4d36f5640b08a73b4c54208b2ed273","abstract_canon_sha256":"4796488e34097585d6dcdb1100edeca82a7db25b712bd730444983d85f63a2c5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:11:04.303452Z","signature_b64":"GkzW9eq1eeRs71F/61XkPXJUV1jmVKilPaQx3KT/srTRft0TUQ6rzx/AYFmK/OUeFGAIlu8mWZzoApi8Z/ORCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9df8dc1a0b668f3b0f23d627c5669bb1c12cc44d4cdde43e6f959737cd926fee","last_reissued_at":"2026-07-05T11:11:04.303048Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:11:04.303048Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SAGE-Eval: Evaluating LLMs for Systematic Generalizations of Safety Facts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Brenden M. Lake, Chen Yueh-Han, Guy Davidson","submitted_at":"2025-05-27T23:29:32Z","abstract_excerpt":"Do LLMs robustly generalize critical safety facts to novel situations? Lacking this ability is dangerous when users ask naive questions. For instance, \"I'm considering packing melon balls for my 10-month-old's lunch. What other foods would be good to include?\" Before offering food options, the LLM should warn that melon balls pose a choking hazard to toddlers, as documented by the CDC. Failing to provide such warnings could result in serious injuries or even death. To evaluate this, we introduce SAGE-Eval, SAfety-fact systematic GEneralization evaluation, the first benchmark that tests whether"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.21828","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2505.21828/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2505.21828","created_at":"2026-07-05T11:11:04.303103+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.21828v1","created_at":"2026-07-05T11:11:04.303103+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.21828","created_at":"2026-07-05T11:11:04.303103+00:00"},{"alias_kind":"pith_short_12","alias_value":"TX4NYGQLM2HT","created_at":"2026-07-05T11:11:04.303103+00:00"},{"alias_kind":"pith_short_16","alias_value":"TX4NYGQLM2HTWDZD","created_at":"2026-07-05T11:11:04.303103+00:00"},{"alias_kind":"pith_short_8","alias_value":"TX4NYGQL","created_at":"2026-07-05T11:11:04.303103+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TX4NYGQLM2HTWDZD2YT4KZU3WH","json":"https://pith.science/pith/TX4NYGQLM2HTWDZD2YT4KZU3WH.json","graph_json":"https://pith.science/api/pith-number/TX4NYGQLM2HTWDZD2YT4KZU3WH/graph.json","events_json":"https://pith.science/api/pith-number/TX4NYGQLM2HTWDZD2YT4KZU3WH/events.json","paper":"https://pith.science/paper/TX4NYGQL"},"agent_actions":{"view_html":"https://pith.science/pith/TX4NYGQLM2HTWDZD2YT4KZU3WH","download_json":"https://pith.science/pith/TX4NYGQLM2HTWDZD2YT4KZU3WH.json","view_paper":"https://pith.science/paper/TX4NYGQL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.21828&json=true","fetch_graph":"https://pith.science/api/pith-number/TX4NYGQLM2HTWDZD2YT4KZU3WH/graph.json","fetch_events":"https://pith.science/api/pith-number/TX4NYGQLM2HTWDZD2YT4KZU3WH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TX4NYGQLM2HTWDZD2YT4KZU3WH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TX4NYGQLM2HTWDZD2YT4KZU3WH/action/storage_attestation","attest_author":"https://pith.science/pith/TX4NYGQLM2HTWDZD2YT4KZU3WH/action/author_attestation","sign_citation":"https://pith.science/pith/TX4NYGQLM2HTWDZD2YT4KZU3WH/action/citation_signature","submit_replication":"https://pith.science/pith/TX4NYGQLM2HTWDZD2YT4KZU3WH/action/replication_record"}},"created_at":"2026-07-05T11:11:04.303103+00:00","updated_at":"2026-07-05T11:11:04.303103+00:00"}