{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:GSGZUFYOMWNXAZZJM2XNJQ7TRX","short_pith_number":"pith:GSGZUFYO","schema_version":"1.0","canonical_sha256":"348d9a170e659b70672966aed4c3f38df0c98cc4590871a53e8cde6871f8042f","source":{"kind":"arxiv","id":"2406.12223","version":1},"attestation_state":"computed","paper":{"title":"ToxiCloakCN: Evaluating Robustness of Offensive Language Detection in Chinese with Cloaking Perturbations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CY"],"primary_cat":"cs.CL","authors_text":"Kenny Tsu Wei Choo, Roy Ka-Wei Lee, Yujia Hu, Yunze Xiao","submitted_at":"2024-06-18T02:44:56Z","abstract_excerpt":"Detecting hate speech and offensive language is essential for maintaining a safe and respectful digital environment. This study examines the limitations of state-of-the-art large language models (LLMs) in identifying offensive content within systematically perturbed data, with a focus on Chinese, a language particularly susceptible to such perturbations. We introduce \\textsf{ToxiCloakCN}, an enhanced dataset derived from ToxiCN, augmented with homophonic substitutions and emoji transformations, to test the robustness of LLMs against these cloaking perturbations. Our findings reveal that existi"},"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":"2406.12223","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-18T02:44:56Z","cross_cats_sorted":["cs.CY"],"title_canon_sha256":"fe39d057b083d53afc6a122ce404808ebd33a636cd22253c81e272593dd2c3cc","abstract_canon_sha256":"5091ec7df41710340adab48d0dcb4d896afbd0c47a59d59237707c2dbbdb28d5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:33:42.793400Z","signature_b64":"shI450OY42I4tvtTPYPJia6/V6TBO76gyWeRw4ZCQHF2HrHIOYC9bQJk2KqkjwklTT134i+chvqY1hS96BwvDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"348d9a170e659b70672966aed4c3f38df0c98cc4590871a53e8cde6871f8042f","last_reissued_at":"2026-07-05T08:33:42.792929Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:33:42.792929Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ToxiCloakCN: Evaluating Robustness of Offensive Language Detection in Chinese with Cloaking Perturbations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CY"],"primary_cat":"cs.CL","authors_text":"Kenny Tsu Wei Choo, Roy Ka-Wei Lee, Yujia Hu, Yunze Xiao","submitted_at":"2024-06-18T02:44:56Z","abstract_excerpt":"Detecting hate speech and offensive language is essential for maintaining a safe and respectful digital environment. This study examines the limitations of state-of-the-art large language models (LLMs) in identifying offensive content within systematically perturbed data, with a focus on Chinese, a language particularly susceptible to such perturbations. We introduce \\textsf{ToxiCloakCN}, an enhanced dataset derived from ToxiCN, augmented with homophonic substitutions and emoji transformations, to test the robustness of LLMs against these cloaking perturbations. Our findings reveal that existi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.12223","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/2406.12223/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":"2406.12223","created_at":"2026-07-05T08:33:42.792985+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.12223v1","created_at":"2026-07-05T08:33:42.792985+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.12223","created_at":"2026-07-05T08:33:42.792985+00:00"},{"alias_kind":"pith_short_12","alias_value":"GSGZUFYOMWNX","created_at":"2026-07-05T08:33:42.792985+00:00"},{"alias_kind":"pith_short_16","alias_value":"GSGZUFYOMWNXAZZJ","created_at":"2026-07-05T08:33:42.792985+00:00"},{"alias_kind":"pith_short_8","alias_value":"GSGZUFYO","created_at":"2026-07-05T08:33:42.792985+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.07640","citing_title":"Lost in Pronunciation: Detecting Chinese Offensive Language Disguised by Phonetic Cloaking Replacement","ref_index":34,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GSGZUFYOMWNXAZZJM2XNJQ7TRX","json":"https://pith.science/pith/GSGZUFYOMWNXAZZJM2XNJQ7TRX.json","graph_json":"https://pith.science/api/pith-number/GSGZUFYOMWNXAZZJM2XNJQ7TRX/graph.json","events_json":"https://pith.science/api/pith-number/GSGZUFYOMWNXAZZJM2XNJQ7TRX/events.json","paper":"https://pith.science/paper/GSGZUFYO"},"agent_actions":{"view_html":"https://pith.science/pith/GSGZUFYOMWNXAZZJM2XNJQ7TRX","download_json":"https://pith.science/pith/GSGZUFYOMWNXAZZJM2XNJQ7TRX.json","view_paper":"https://pith.science/paper/GSGZUFYO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.12223&json=true","fetch_graph":"https://pith.science/api/pith-number/GSGZUFYOMWNXAZZJM2XNJQ7TRX/graph.json","fetch_events":"https://pith.science/api/pith-number/GSGZUFYOMWNXAZZJM2XNJQ7TRX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GSGZUFYOMWNXAZZJM2XNJQ7TRX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GSGZUFYOMWNXAZZJM2XNJQ7TRX/action/storage_attestation","attest_author":"https://pith.science/pith/GSGZUFYOMWNXAZZJM2XNJQ7TRX/action/author_attestation","sign_citation":"https://pith.science/pith/GSGZUFYOMWNXAZZJM2XNJQ7TRX/action/citation_signature","submit_replication":"https://pith.science/pith/GSGZUFYOMWNXAZZJM2XNJQ7TRX/action/replication_record"}},"created_at":"2026-07-05T08:33:42.792985+00:00","updated_at":"2026-07-05T08:33:42.792985+00:00"}