{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:7MDB46VNYEY25RR2CQ2LIPORRX","short_pith_number":"pith:7MDB46VN","schema_version":"1.0","canonical_sha256":"fb061e7aadc131aec63a1434b43dd18df871efc7f0ea746910a9c40f325246a6","source":{"kind":"arxiv","id":"2405.09373","version":3},"attestation_state":"computed","paper":{"title":"PolygloToxicityPrompts: Multilingual Evaluation of Neural Toxic Degeneration in Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Devansh Jain, Maarten Sap, Priyanshu Kumar, Samuel Gehman, Thomas Hartvigsen, Xuhui Zhou","submitted_at":"2024-05-15T14:22:33Z","abstract_excerpt":"Recent advances in large language models (LLMs) have led to their extensive global deployment, and ensuring their safety calls for comprehensive and multilingual toxicity evaluations. However, existing toxicity benchmarks are overwhelmingly focused on English, posing serious risks to deploying LLMs in other languages. We address this by introducing PolygloToxicityPrompts (PTP), the first large-scale multilingual toxicity evaluation benchmark of 425K naturally occurring prompts spanning 17 languages. We overcome the scarcity of naturally occurring toxicity in web-text and ensure coverage across"},"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":"2405.09373","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-15T14:22:33Z","cross_cats_sorted":[],"title_canon_sha256":"1c15fb9621c92dc177a92f658c78796083272c5ae73a94276ab34ef989a2bdbc","abstract_canon_sha256":"ba29bf9bf421faf70a4356302fb1e6f0602e67782a3d8325460096f663893947"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:54:13.348755Z","signature_b64":"+hmKiyzy8qWHRStqOrTZjDCxOW/4Jv8jUjgdWtEMfMtS+Rh9LQo5cKWmKdoyVC4EIXYhYSs+daOYoe4/pt62Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fb061e7aadc131aec63a1434b43dd18df871efc7f0ea746910a9c40f325246a6","last_reissued_at":"2026-07-05T08:54:13.348258Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:54:13.348258Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PolygloToxicityPrompts: Multilingual Evaluation of Neural Toxic Degeneration in Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Devansh Jain, Maarten Sap, Priyanshu Kumar, Samuel Gehman, Thomas Hartvigsen, Xuhui Zhou","submitted_at":"2024-05-15T14:22:33Z","abstract_excerpt":"Recent advances in large language models (LLMs) have led to their extensive global deployment, and ensuring their safety calls for comprehensive and multilingual toxicity evaluations. However, existing toxicity benchmarks are overwhelmingly focused on English, posing serious risks to deploying LLMs in other languages. We address this by introducing PolygloToxicityPrompts (PTP), the first large-scale multilingual toxicity evaluation benchmark of 425K naturally occurring prompts spanning 17 languages. We overcome the scarcity of naturally occurring toxicity in web-text and ensure coverage across"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.09373","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/2405.09373/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":"2405.09373","created_at":"2026-07-05T08:54:13.348319+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.09373v3","created_at":"2026-07-05T08:54:13.348319+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.09373","created_at":"2026-07-05T08:54:13.348319+00:00"},{"alias_kind":"pith_short_12","alias_value":"7MDB46VNYEY2","created_at":"2026-07-05T08:54:13.348319+00:00"},{"alias_kind":"pith_short_16","alias_value":"7MDB46VNYEY25RR2","created_at":"2026-07-05T08:54:13.348319+00:00"},{"alias_kind":"pith_short_8","alias_value":"7MDB46VN","created_at":"2026-07-05T08:54:13.348319+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.27997","citing_title":"Where Does Toxicity Live? Mechanistic Localization and Targeted Suppression in Language Models","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2605.29659","citing_title":"Opir: Efficient Multi-Task Safety Classification for Toxicity, Jailbreaks, Hate Speech, and Harmful Content","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14087","citing_title":"Measuring and Mitigating Toxicity in Large Language Models: A Comprehensive Replication Study","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14087","citing_title":"Measuring and Mitigating Toxicity in Large Language Models: A Comprehensive Replication Study","ref_index":22,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7MDB46VNYEY25RR2CQ2LIPORRX","json":"https://pith.science/pith/7MDB46VNYEY25RR2CQ2LIPORRX.json","graph_json":"https://pith.science/api/pith-number/7MDB46VNYEY25RR2CQ2LIPORRX/graph.json","events_json":"https://pith.science/api/pith-number/7MDB46VNYEY25RR2CQ2LIPORRX/events.json","paper":"https://pith.science/paper/7MDB46VN"},"agent_actions":{"view_html":"https://pith.science/pith/7MDB46VNYEY25RR2CQ2LIPORRX","download_json":"https://pith.science/pith/7MDB46VNYEY25RR2CQ2LIPORRX.json","view_paper":"https://pith.science/paper/7MDB46VN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.09373&json=true","fetch_graph":"https://pith.science/api/pith-number/7MDB46VNYEY25RR2CQ2LIPORRX/graph.json","fetch_events":"https://pith.science/api/pith-number/7MDB46VNYEY25RR2CQ2LIPORRX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7MDB46VNYEY25RR2CQ2LIPORRX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7MDB46VNYEY25RR2CQ2LIPORRX/action/storage_attestation","attest_author":"https://pith.science/pith/7MDB46VNYEY25RR2CQ2LIPORRX/action/author_attestation","sign_citation":"https://pith.science/pith/7MDB46VNYEY25RR2CQ2LIPORRX/action/citation_signature","submit_replication":"https://pith.science/pith/7MDB46VNYEY25RR2CQ2LIPORRX/action/replication_record"}},"created_at":"2026-07-05T08:54:13.348319+00:00","updated_at":"2026-07-05T08:54:13.348319+00:00"}