{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:PXRBL7W4QWEMSOC4CIIHRIO5OW","short_pith_number":"pith:PXRBL7W4","schema_version":"1.0","canonical_sha256":"7de215fedc8588c9385c121078a1dd75bde694bd57b6800634e59f99a00739f0","source":{"kind":"arxiv","id":"2504.16120","version":1},"attestation_state":"computed","paper":{"title":"A Data-Centric Approach for Safe and Secure Large Language Models against Threatening and Toxic Content","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Chaima Njeh, Fehmi Jaafar, Ha\\\"ifa Nakouri","submitted_at":"2025-04-19T04:57:05Z","abstract_excerpt":"Large Language Models (LLM) have made remarkable progress, but concerns about potential biases and harmful content persist. To address these apprehensions, we introduce a practical solution for ensuring LLM's safe and ethical use. Our novel approach focuses on a post-generation correction mechanism, the BART-Corrective Model, which adjusts generated content to ensure safety and security. Unlike relying solely on model fine-tuning or prompt engineering, our method provides a robust data-centric alternative for mitigating harmful content. We demonstrate the effectiveness of our approach through "},"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":"2504.16120","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-04-19T04:57:05Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"cf4f9b893e5d19e55a58b6f7dc94fa5ffd00a841978dac0e30953d6e7b1bf985","abstract_canon_sha256":"5b3766699287da2dc907b43bbcbc2a119a60cac4d833d03e67f27b9ab733232a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:52:41.634019Z","signature_b64":"Q3UUK+TKrz7Fchpo+aQWNzMFN6r/vSu5+5QXY9119/UfFJWQPpO9pOekxrMe/vktCYGnzQw9BX9+XN5dF2BZBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7de215fedc8588c9385c121078a1dd75bde694bd57b6800634e59f99a00739f0","last_reissued_at":"2026-07-05T10:52:41.633585Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:52:41.633585Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Data-Centric Approach for Safe and Secure Large Language Models against Threatening and Toxic Content","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Chaima Njeh, Fehmi Jaafar, Ha\\\"ifa Nakouri","submitted_at":"2025-04-19T04:57:05Z","abstract_excerpt":"Large Language Models (LLM) have made remarkable progress, but concerns about potential biases and harmful content persist. To address these apprehensions, we introduce a practical solution for ensuring LLM's safe and ethical use. Our novel approach focuses on a post-generation correction mechanism, the BART-Corrective Model, which adjusts generated content to ensure safety and security. Unlike relying solely on model fine-tuning or prompt engineering, our method provides a robust data-centric alternative for mitigating harmful content. We demonstrate the effectiveness of our approach through "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.16120","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/2504.16120/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":"2504.16120","created_at":"2026-07-05T10:52:41.633634+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.16120v1","created_at":"2026-07-05T10:52:41.633634+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.16120","created_at":"2026-07-05T10:52:41.633634+00:00"},{"alias_kind":"pith_short_12","alias_value":"PXRBL7W4QWEM","created_at":"2026-07-05T10:52:41.633634+00:00"},{"alias_kind":"pith_short_16","alias_value":"PXRBL7W4QWEMSOC4","created_at":"2026-07-05T10:52:41.633634+00:00"},{"alias_kind":"pith_short_8","alias_value":"PXRBL7W4","created_at":"2026-07-05T10:52:41.633634+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PXRBL7W4QWEMSOC4CIIHRIO5OW","json":"https://pith.science/pith/PXRBL7W4QWEMSOC4CIIHRIO5OW.json","graph_json":"https://pith.science/api/pith-number/PXRBL7W4QWEMSOC4CIIHRIO5OW/graph.json","events_json":"https://pith.science/api/pith-number/PXRBL7W4QWEMSOC4CIIHRIO5OW/events.json","paper":"https://pith.science/paper/PXRBL7W4"},"agent_actions":{"view_html":"https://pith.science/pith/PXRBL7W4QWEMSOC4CIIHRIO5OW","download_json":"https://pith.science/pith/PXRBL7W4QWEMSOC4CIIHRIO5OW.json","view_paper":"https://pith.science/paper/PXRBL7W4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.16120&json=true","fetch_graph":"https://pith.science/api/pith-number/PXRBL7W4QWEMSOC4CIIHRIO5OW/graph.json","fetch_events":"https://pith.science/api/pith-number/PXRBL7W4QWEMSOC4CIIHRIO5OW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PXRBL7W4QWEMSOC4CIIHRIO5OW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PXRBL7W4QWEMSOC4CIIHRIO5OW/action/storage_attestation","attest_author":"https://pith.science/pith/PXRBL7W4QWEMSOC4CIIHRIO5OW/action/author_attestation","sign_citation":"https://pith.science/pith/PXRBL7W4QWEMSOC4CIIHRIO5OW/action/citation_signature","submit_replication":"https://pith.science/pith/PXRBL7W4QWEMSOC4CIIHRIO5OW/action/replication_record"}},"created_at":"2026-07-05T10:52:41.633634+00:00","updated_at":"2026-07-05T10:52:41.633634+00:00"}