{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:ZOTQKDXTMP2XXG4LTCWDCZIDHA","short_pith_number":"pith:ZOTQKDXT","schema_version":"1.0","canonical_sha256":"cba7050ef363f57b9b8b98ac3165033835951bff857936f168e1ca32f31fdaa4","source":{"kind":"arxiv","id":"2506.10960","version":3},"attestation_state":"computed","paper":{"title":"ChineseHarm-Bench: A Chinese Harmful Content Detection Benchmark","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CR","cs.IR","cs.LG"],"primary_cat":"cs.CL","authors_text":"Bozhong Tian, Bryan Hooi, Kangwei Liu, Meng Han, Ningyu Zhang, Shumin Deng, Siyuan Cheng, Xiaozhuan Liang, Xi Chen, Yuyang Yin","submitted_at":"2025-06-12T17:57:05Z","abstract_excerpt":"Large language models (LLMs) have been increasingly applied to automated harmful content detection tasks, assisting moderators in identifying policy violations and improving the overall efficiency and accuracy of content review. However, existing resources for harmful content detection are predominantly focused on English, with Chinese datasets remaining scarce and often limited in scope. We present a comprehensive, professionally annotated benchmark for Chinese content harm detection, which covers six representative categories and is constructed entirely from real-world data. Our annotation p"},"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":"2506.10960","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-12T17:57:05Z","cross_cats_sorted":["cs.AI","cs.CR","cs.IR","cs.LG"],"title_canon_sha256":"7147939b323ec6c6fd63fd61029b55a0a1cffa5b86cc9bf17db173598dfc2ddc","abstract_canon_sha256":"d29d85c702f49702dc63bafa43bb9d8a93e1ebc77b6bf16d28de58e3733acc99"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:53:04.196160Z","signature_b64":"Oa5m6KQFWKFGHNg5MrKGNqSZkAFLPbwHEq6mMemhnC/d+SsNTms8g0bwtfpGktoSYtTPZ9nhdIAm6+UxU8TLDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cba7050ef363f57b9b8b98ac3165033835951bff857936f168e1ca32f31fdaa4","last_reissued_at":"2026-07-05T11:53:04.195633Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:53:04.195633Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ChineseHarm-Bench: A Chinese Harmful Content Detection Benchmark","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CR","cs.IR","cs.LG"],"primary_cat":"cs.CL","authors_text":"Bozhong Tian, Bryan Hooi, Kangwei Liu, Meng Han, Ningyu Zhang, Shumin Deng, Siyuan Cheng, Xiaozhuan Liang, Xi Chen, Yuyang Yin","submitted_at":"2025-06-12T17:57:05Z","abstract_excerpt":"Large language models (LLMs) have been increasingly applied to automated harmful content detection tasks, assisting moderators in identifying policy violations and improving the overall efficiency and accuracy of content review. However, existing resources for harmful content detection are predominantly focused on English, with Chinese datasets remaining scarce and often limited in scope. We present a comprehensive, professionally annotated benchmark for Chinese content harm detection, which covers six representative categories and is constructed entirely from real-world data. Our annotation p"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.10960","kind":"arxiv","version":3},"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/2506.10960/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":"2506.10960","created_at":"2026-07-05T11:53:04.195703+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.10960v3","created_at":"2026-07-05T11:53:04.195703+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.10960","created_at":"2026-07-05T11:53:04.195703+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZOTQKDXTMP2X","created_at":"2026-07-05T11:53:04.195703+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZOTQKDXTMP2XXG4L","created_at":"2026-07-05T11:53:04.195703+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZOTQKDXT","created_at":"2026-07-05T11:53:04.195703+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.27632","citing_title":"Yuvion LLM: An Adversarially-Aware Large Language Model for Content And AI Safety","ref_index":20,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZOTQKDXTMP2XXG4LTCWDCZIDHA","json":"https://pith.science/pith/ZOTQKDXTMP2XXG4LTCWDCZIDHA.json","graph_json":"https://pith.science/api/pith-number/ZOTQKDXTMP2XXG4LTCWDCZIDHA/graph.json","events_json":"https://pith.science/api/pith-number/ZOTQKDXTMP2XXG4LTCWDCZIDHA/events.json","paper":"https://pith.science/paper/ZOTQKDXT"},"agent_actions":{"view_html":"https://pith.science/pith/ZOTQKDXTMP2XXG4LTCWDCZIDHA","download_json":"https://pith.science/pith/ZOTQKDXTMP2XXG4LTCWDCZIDHA.json","view_paper":"https://pith.science/paper/ZOTQKDXT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.10960&json=true","fetch_graph":"https://pith.science/api/pith-number/ZOTQKDXTMP2XXG4LTCWDCZIDHA/graph.json","fetch_events":"https://pith.science/api/pith-number/ZOTQKDXTMP2XXG4LTCWDCZIDHA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZOTQKDXTMP2XXG4LTCWDCZIDHA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZOTQKDXTMP2XXG4LTCWDCZIDHA/action/storage_attestation","attest_author":"https://pith.science/pith/ZOTQKDXTMP2XXG4LTCWDCZIDHA/action/author_attestation","sign_citation":"https://pith.science/pith/ZOTQKDXTMP2XXG4LTCWDCZIDHA/action/citation_signature","submit_replication":"https://pith.science/pith/ZOTQKDXTMP2XXG4LTCWDCZIDHA/action/replication_record"}},"created_at":"2026-07-05T11:53:04.195703+00:00","updated_at":"2026-07-05T11:53:04.195703+00:00"}