{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:UAXPJ7Y7LXJRDXZITC65PWQR4S","short_pith_number":"pith:UAXPJ7Y7","schema_version":"1.0","canonical_sha256":"a02ef4ff1f5dd311df2898bdd7da11e4876d65cc62849cd27882c5f215e5cb36","source":{"kind":"arxiv","id":"2505.18927","version":3},"attestation_state":"computed","paper":{"title":"Moderating Harm: Benchmarking Large Language Models for Cyberbullying Detection in YouTube Comments","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Amel Muminovic","submitted_at":"2025-05-25T01:28:30Z","abstract_excerpt":"As online platforms grow, comment sections increasingly host harassment that undermines user experience and well-being. This study benchmarks three leading large language models, OpenAI GPT-4.1, Google Gemini 1.5 Pro, and Anthropic Claude 3 Opus, on a corpus of 5,080 YouTube comments sampled from high-abuse threads in gaming, lifestyle, food vlog, and music channels. The dataset comprises 1,334 harmful and 3,746 non-harmful messages in English, Arabic, and Indonesian, annotated independently by two reviewers with substantial agreement (Cohen's kappa = 0.83). Using a unified prompt and determin"},"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":"2505.18927","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-25T01:28:30Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"7d472aa57b02c15a99329e61cb49da99a41d14d643298b1616a2773d63aa7485","abstract_canon_sha256":"aab0c67850a1bd4046513d9ad0fe071d252d4c37927a2a92bdf5a44a7ca3d919"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:13:26.652277Z","signature_b64":"tglRTHPIl4KSEZNZxp212MARIORS+f5AzCHswvb9ORWf9MJ9FmdNCMv8mFy5mCuykpIjOAHW7uSKbhpsQykIDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a02ef4ff1f5dd311df2898bdd7da11e4876d65cc62849cd27882c5f215e5cb36","last_reissued_at":"2026-07-05T11:13:26.651848Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:13:26.651848Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Moderating Harm: Benchmarking Large Language Models for Cyberbullying Detection in YouTube Comments","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Amel Muminovic","submitted_at":"2025-05-25T01:28:30Z","abstract_excerpt":"As online platforms grow, comment sections increasingly host harassment that undermines user experience and well-being. This study benchmarks three leading large language models, OpenAI GPT-4.1, Google Gemini 1.5 Pro, and Anthropic Claude 3 Opus, on a corpus of 5,080 YouTube comments sampled from high-abuse threads in gaming, lifestyle, food vlog, and music channels. The dataset comprises 1,334 harmful and 3,746 non-harmful messages in English, Arabic, and Indonesian, annotated independently by two reviewers with substantial agreement (Cohen's kappa = 0.83). Using a unified prompt and determin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.18927","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/2505.18927/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":"2505.18927","created_at":"2026-07-05T11:13:26.651905+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.18927v3","created_at":"2026-07-05T11:13:26.651905+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.18927","created_at":"2026-07-05T11:13:26.651905+00:00"},{"alias_kind":"pith_short_12","alias_value":"UAXPJ7Y7LXJR","created_at":"2026-07-05T11:13:26.651905+00:00"},{"alias_kind":"pith_short_16","alias_value":"UAXPJ7Y7LXJRDXZI","created_at":"2026-07-05T11:13:26.651905+00:00"},{"alias_kind":"pith_short_8","alias_value":"UAXPJ7Y7","created_at":"2026-07-05T11:13:26.651905+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.27584","citing_title":"Cyberbullying Governance on Social Media: A Unified Framework from Content Identification to Intervention","ref_index":151,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UAXPJ7Y7LXJRDXZITC65PWQR4S","json":"https://pith.science/pith/UAXPJ7Y7LXJRDXZITC65PWQR4S.json","graph_json":"https://pith.science/api/pith-number/UAXPJ7Y7LXJRDXZITC65PWQR4S/graph.json","events_json":"https://pith.science/api/pith-number/UAXPJ7Y7LXJRDXZITC65PWQR4S/events.json","paper":"https://pith.science/paper/UAXPJ7Y7"},"agent_actions":{"view_html":"https://pith.science/pith/UAXPJ7Y7LXJRDXZITC65PWQR4S","download_json":"https://pith.science/pith/UAXPJ7Y7LXJRDXZITC65PWQR4S.json","view_paper":"https://pith.science/paper/UAXPJ7Y7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.18927&json=true","fetch_graph":"https://pith.science/api/pith-number/UAXPJ7Y7LXJRDXZITC65PWQR4S/graph.json","fetch_events":"https://pith.science/api/pith-number/UAXPJ7Y7LXJRDXZITC65PWQR4S/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UAXPJ7Y7LXJRDXZITC65PWQR4S/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UAXPJ7Y7LXJRDXZITC65PWQR4S/action/storage_attestation","attest_author":"https://pith.science/pith/UAXPJ7Y7LXJRDXZITC65PWQR4S/action/author_attestation","sign_citation":"https://pith.science/pith/UAXPJ7Y7LXJRDXZITC65PWQR4S/action/citation_signature","submit_replication":"https://pith.science/pith/UAXPJ7Y7LXJRDXZITC65PWQR4S/action/replication_record"}},"created_at":"2026-07-05T11:13:26.651905+00:00","updated_at":"2026-07-05T11:13:26.651905+00:00"}