{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:NF7DU3LZKMBN5W37V6HOKV2AFJ","short_pith_number":"pith:NF7DU3LZ","schema_version":"1.0","canonical_sha256":"697e3a6d795302dedb7faf8ee557402a48462c26e64ba142e8e537540a5c8acc","source":{"kind":"arxiv","id":"2503.22115","version":1},"attestation_state":"computed","paper":{"title":"Beyond Single-Sentence Prompts: Upgrading Value Alignment Benchmarks with Dialogues and Stories","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CY"],"primary_cat":"cs.CL","authors_text":"Jing Qin, Peng Zhang, Qimeng Liu, Qiuchi Li, Yazhou Zhang","submitted_at":"2025-03-28T03:31:37Z","abstract_excerpt":"Evaluating the value alignment of large language models (LLMs) has traditionally relied on single-sentence adversarial prompts, which directly probe models with ethically sensitive or controversial questions. However, with the rapid advancements in AI safety techniques, models have become increasingly adept at circumventing these straightforward tests, limiting their effectiveness in revealing underlying biases and ethical stances. To address this limitation, we propose an upgraded value alignment benchmark that moves beyond single-sentence prompts by incorporating multi-turn dialogues and nar"},"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":"2503.22115","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-03-28T03:31:37Z","cross_cats_sorted":["cs.AI","cs.CY"],"title_canon_sha256":"58cd50d5b1ee93da2e09dea817d4654f7a20687331584034798bec0907b611ab","abstract_canon_sha256":"3ad50656f90a139e8b6fac2c0e2901797f2e6c01061f8357d70aeacd16257b12"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:40:47.553437Z","signature_b64":"AP+wbd1pbXmbLRaqGaXGx8w2hWcZcn5w14uszzBids/bMbxAHeiDd+Py69dHrxAqYabGOxTMkGGIrHNx2kodAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"697e3a6d795302dedb7faf8ee557402a48462c26e64ba142e8e537540a5c8acc","last_reissued_at":"2026-07-05T10:40:47.552951Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:40:47.552951Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Beyond Single-Sentence Prompts: Upgrading Value Alignment Benchmarks with Dialogues and Stories","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CY"],"primary_cat":"cs.CL","authors_text":"Jing Qin, Peng Zhang, Qimeng Liu, Qiuchi Li, Yazhou Zhang","submitted_at":"2025-03-28T03:31:37Z","abstract_excerpt":"Evaluating the value alignment of large language models (LLMs) has traditionally relied on single-sentence adversarial prompts, which directly probe models with ethically sensitive or controversial questions. However, with the rapid advancements in AI safety techniques, models have become increasingly adept at circumventing these straightforward tests, limiting their effectiveness in revealing underlying biases and ethical stances. To address this limitation, we propose an upgraded value alignment benchmark that moves beyond single-sentence prompts by incorporating multi-turn dialogues and nar"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.22115","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/2503.22115/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":"2503.22115","created_at":"2026-07-05T10:40:47.553014+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.22115v1","created_at":"2026-07-05T10:40:47.553014+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.22115","created_at":"2026-07-05T10:40:47.553014+00:00"},{"alias_kind":"pith_short_12","alias_value":"NF7DU3LZKMBN","created_at":"2026-07-05T10:40:47.553014+00:00"},{"alias_kind":"pith_short_16","alias_value":"NF7DU3LZKMBN5W37","created_at":"2026-07-05T10:40:47.553014+00:00"},{"alias_kind":"pith_short_8","alias_value":"NF7DU3LZ","created_at":"2026-07-05T10:40:47.553014+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/NF7DU3LZKMBN5W37V6HOKV2AFJ","json":"https://pith.science/pith/NF7DU3LZKMBN5W37V6HOKV2AFJ.json","graph_json":"https://pith.science/api/pith-number/NF7DU3LZKMBN5W37V6HOKV2AFJ/graph.json","events_json":"https://pith.science/api/pith-number/NF7DU3LZKMBN5W37V6HOKV2AFJ/events.json","paper":"https://pith.science/paper/NF7DU3LZ"},"agent_actions":{"view_html":"https://pith.science/pith/NF7DU3LZKMBN5W37V6HOKV2AFJ","download_json":"https://pith.science/pith/NF7DU3LZKMBN5W37V6HOKV2AFJ.json","view_paper":"https://pith.science/paper/NF7DU3LZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.22115&json=true","fetch_graph":"https://pith.science/api/pith-number/NF7DU3LZKMBN5W37V6HOKV2AFJ/graph.json","fetch_events":"https://pith.science/api/pith-number/NF7DU3LZKMBN5W37V6HOKV2AFJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NF7DU3LZKMBN5W37V6HOKV2AFJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NF7DU3LZKMBN5W37V6HOKV2AFJ/action/storage_attestation","attest_author":"https://pith.science/pith/NF7DU3LZKMBN5W37V6HOKV2AFJ/action/author_attestation","sign_citation":"https://pith.science/pith/NF7DU3LZKMBN5W37V6HOKV2AFJ/action/citation_signature","submit_replication":"https://pith.science/pith/NF7DU3LZKMBN5W37V6HOKV2AFJ/action/replication_record"}},"created_at":"2026-07-05T10:40:47.553014+00:00","updated_at":"2026-07-05T10:40:47.553014+00:00"}