{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:BOY2NPJVSJ3OGDM3YOVYKO3GHQ","short_pith_number":"pith:BOY2NPJV","schema_version":"1.0","canonical_sha256":"0bb1a6bd359276e30d9bc3ab853b663c05321c06bd6748be0a151a8a9dd4e9af","source":{"kind":"arxiv","id":"2505.18154","version":1},"attestation_state":"computed","paper":{"title":"The Staircase of Ethics: Probing LLM Value Priorities through Multi-Step Induction to Complex Moral Dilemmas","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CY"],"primary_cat":"cs.CL","authors_text":"Danding Wang, Guang Yang, Juan Cao, Qiang Sheng, Ya Wu, Yifan Sun, Yuyan Bu, Zhengjia Wang","submitted_at":"2025-05-23T17:59:50Z","abstract_excerpt":"Ethical decision-making is a critical aspect of human judgment, and the growing use of LLMs in decision-support systems necessitates a rigorous evaluation of their moral reasoning capabilities. However, existing assessments primarily rely on single-step evaluations, failing to capture how models adapt to evolving ethical challenges. Addressing this gap, we introduce the Multi-step Moral Dilemmas (MMDs), the first dataset specifically constructed to evaluate the evolving moral judgments of LLMs across 3,302 five-stage dilemmas. This framework enables a fine-grained, dynamic analysis of how LLMs"},"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":"2505.18154","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-23T17:59:50Z","cross_cats_sorted":["cs.CY"],"title_canon_sha256":"c5650eafa6dcedc1da717ef0dbbe452c2da6dad49e7ada4a42d9a302e1ff3e10","abstract_canon_sha256":"bcb6f9587469b13cc9c9928ae572af2423a9c0733cfdf9a5558c5549d4ed5101"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:08:37.191726Z","signature_b64":"uHYWH2nur7o2Rd4TZwfKXMUrA7moALXoH5rXsFO7tjfZMiXziE+68wu1+84BO0Fe9BFpET181ZleaV+GVp6TDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0bb1a6bd359276e30d9bc3ab853b663c05321c06bd6748be0a151a8a9dd4e9af","last_reissued_at":"2026-07-05T11:08:37.191247Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:08:37.191247Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Staircase of Ethics: Probing LLM Value Priorities through Multi-Step Induction to Complex Moral Dilemmas","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CY"],"primary_cat":"cs.CL","authors_text":"Danding Wang, Guang Yang, Juan Cao, Qiang Sheng, Ya Wu, Yifan Sun, Yuyan Bu, Zhengjia Wang","submitted_at":"2025-05-23T17:59:50Z","abstract_excerpt":"Ethical decision-making is a critical aspect of human judgment, and the growing use of LLMs in decision-support systems necessitates a rigorous evaluation of their moral reasoning capabilities. However, existing assessments primarily rely on single-step evaluations, failing to capture how models adapt to evolving ethical challenges. Addressing this gap, we introduce the Multi-step Moral Dilemmas (MMDs), the first dataset specifically constructed to evaluate the evolving moral judgments of LLMs across 3,302 five-stage dilemmas. This framework enables a fine-grained, dynamic analysis of how LLMs"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.18154","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/2505.18154/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":"2505.18154","created_at":"2026-07-05T11:08:37.191304+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.18154v1","created_at":"2026-07-05T11:08:37.191304+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.18154","created_at":"2026-07-05T11:08:37.191304+00:00"},{"alias_kind":"pith_short_12","alias_value":"BOY2NPJVSJ3O","created_at":"2026-07-05T11:08:37.191304+00:00"},{"alias_kind":"pith_short_16","alias_value":"BOY2NPJVSJ3OGDM3","created_at":"2026-07-05T11:08:37.191304+00:00"},{"alias_kind":"pith_short_8","alias_value":"BOY2NPJV","created_at":"2026-07-05T11:08:37.191304+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.02214","citing_title":"Do Gender Cues Affect LLM Value Trade-offs? Evidence from a Controlled Decision Benchmark","ref_index":26,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BOY2NPJVSJ3OGDM3YOVYKO3GHQ","json":"https://pith.science/pith/BOY2NPJVSJ3OGDM3YOVYKO3GHQ.json","graph_json":"https://pith.science/api/pith-number/BOY2NPJVSJ3OGDM3YOVYKO3GHQ/graph.json","events_json":"https://pith.science/api/pith-number/BOY2NPJVSJ3OGDM3YOVYKO3GHQ/events.json","paper":"https://pith.science/paper/BOY2NPJV"},"agent_actions":{"view_html":"https://pith.science/pith/BOY2NPJVSJ3OGDM3YOVYKO3GHQ","download_json":"https://pith.science/pith/BOY2NPJVSJ3OGDM3YOVYKO3GHQ.json","view_paper":"https://pith.science/paper/BOY2NPJV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.18154&json=true","fetch_graph":"https://pith.science/api/pith-number/BOY2NPJVSJ3OGDM3YOVYKO3GHQ/graph.json","fetch_events":"https://pith.science/api/pith-number/BOY2NPJVSJ3OGDM3YOVYKO3GHQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BOY2NPJVSJ3OGDM3YOVYKO3GHQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BOY2NPJVSJ3OGDM3YOVYKO3GHQ/action/storage_attestation","attest_author":"https://pith.science/pith/BOY2NPJVSJ3OGDM3YOVYKO3GHQ/action/author_attestation","sign_citation":"https://pith.science/pith/BOY2NPJVSJ3OGDM3YOVYKO3GHQ/action/citation_signature","submit_replication":"https://pith.science/pith/BOY2NPJVSJ3OGDM3YOVYKO3GHQ/action/replication_record"}},"created_at":"2026-07-05T11:08:37.191304+00:00","updated_at":"2026-07-05T11:08:37.191304+00:00"}