{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:SIDUSORETCLFGXXTOAKTLDII66","short_pith_number":"pith:SIDUSORE","schema_version":"1.0","canonical_sha256":"9207493a249896535ef37015358d08f7be3e262ab31524891413b405b048c5e6","source":{"kind":"arxiv","id":"2505.15365","version":1},"attestation_state":"computed","paper":{"title":"AI vs. Human Judgment of Content Moderation: LLM-as-a-Judge and Ethics-Based Response Refusals","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.HC","authors_text":"Stefan Pasch","submitted_at":"2025-05-21T10:56:16Z","abstract_excerpt":"As large language models (LLMs) are increasingly deployed in high-stakes settings, their ability to refuse ethically sensitive prompts-such as those involving hate speech or illegal activities-has become central to content moderation and responsible AI practices. While refusal responses can be viewed as evidence of ethical alignment and safety-conscious behavior, recent research suggests that users may perceive them negatively. At the same time, automated assessments of model outputs are playing a growing role in both evaluation and training. In particular, LLM-as-a-Judge frameworks-in which o"},"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":"2505.15365","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.HC","submitted_at":"2025-05-21T10:56:16Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"b729190ca1718248ef46b92b37c167bd6f550d2e3fc7909074082ee84ba8bf4d","abstract_canon_sha256":"92c856c173e8db7a34331bd57e12fca46be3d6bb6866d7fbb8081796a322fd63"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:06:44.306060Z","signature_b64":"b1wZ9HhADz/Tk0DcuuMM/t4k3T07ITVxB3G4RBi169gbgZ7+tbQIM7q5y1BlXzbzJjHWxyQvDjpr/7zhYS/QCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9207493a249896535ef37015358d08f7be3e262ab31524891413b405b048c5e6","last_reissued_at":"2026-07-05T11:06:44.305559Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:06:44.305559Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"AI vs. Human Judgment of Content Moderation: LLM-as-a-Judge and Ethics-Based Response Refusals","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.HC","authors_text":"Stefan Pasch","submitted_at":"2025-05-21T10:56:16Z","abstract_excerpt":"As large language models (LLMs) are increasingly deployed in high-stakes settings, their ability to refuse ethically sensitive prompts-such as those involving hate speech or illegal activities-has become central to content moderation and responsible AI practices. While refusal responses can be viewed as evidence of ethical alignment and safety-conscious behavior, recent research suggests that users may perceive them negatively. At the same time, automated assessments of model outputs are playing a growing role in both evaluation and training. In particular, LLM-as-a-Judge frameworks-in which o"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.15365","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/2505.15365/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":"2505.15365","created_at":"2026-07-05T11:06:44.305622+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.15365v1","created_at":"2026-07-05T11:06:44.305622+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.15365","created_at":"2026-07-05T11:06:44.305622+00:00"},{"alias_kind":"pith_short_12","alias_value":"SIDUSORETCLF","created_at":"2026-07-05T11:06:44.305622+00:00"},{"alias_kind":"pith_short_16","alias_value":"SIDUSORETCLFGXXT","created_at":"2026-07-05T11:06:44.305622+00:00"},{"alias_kind":"pith_short_8","alias_value":"SIDUSORE","created_at":"2026-07-05T11:06:44.305622+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2510.15746","citing_title":"LLMs Judge Themselves: A Game-Theoretic Framework for Human-Aligned Evaluation","ref_index":2,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SIDUSORETCLFGXXTOAKTLDII66","json":"https://pith.science/pith/SIDUSORETCLFGXXTOAKTLDII66.json","graph_json":"https://pith.science/api/pith-number/SIDUSORETCLFGXXTOAKTLDII66/graph.json","events_json":"https://pith.science/api/pith-number/SIDUSORETCLFGXXTOAKTLDII66/events.json","paper":"https://pith.science/paper/SIDUSORE"},"agent_actions":{"view_html":"https://pith.science/pith/SIDUSORETCLFGXXTOAKTLDII66","download_json":"https://pith.science/pith/SIDUSORETCLFGXXTOAKTLDII66.json","view_paper":"https://pith.science/paper/SIDUSORE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.15365&json=true","fetch_graph":"https://pith.science/api/pith-number/SIDUSORETCLFGXXTOAKTLDII66/graph.json","fetch_events":"https://pith.science/api/pith-number/SIDUSORETCLFGXXTOAKTLDII66/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SIDUSORETCLFGXXTOAKTLDII66/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SIDUSORETCLFGXXTOAKTLDII66/action/storage_attestation","attest_author":"https://pith.science/pith/SIDUSORETCLFGXXTOAKTLDII66/action/author_attestation","sign_citation":"https://pith.science/pith/SIDUSORETCLFGXXTOAKTLDII66/action/citation_signature","submit_replication":"https://pith.science/pith/SIDUSORETCLFGXXTOAKTLDII66/action/replication_record"}},"created_at":"2026-07-05T11:06:44.305622+00:00","updated_at":"2026-07-05T11:06:44.305622+00:00"}