{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:DN4LQ5YAOV3WLUXW2PAZ4MHB5Z","short_pith_number":"pith:DN4LQ5YA","schema_version":"1.0","canonical_sha256":"1b78b87700757765d2f6d3c19e30e1ee5bc79a89c5df8b644f9331f207a34512","source":{"kind":"arxiv","id":"2508.05775","version":2},"attestation_state":"computed","paper":{"title":"Guardians and Offenders: A Survey on Harmful Content Generation and Safety Mitigation of LLM","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CY"],"primary_cat":"cs.CL","authors_text":"Changjia Zhu, Chi Zhang, Junjie Xiong, Lingyao Li, Xiaoran Xu, Yao Liu, Zhuo Lu","submitted_at":"2025-08-07T18:42:16Z","abstract_excerpt":"Large Language Models (LLMs) have revolutionized content creation across digital platforms, offering unprecedented capabilities in natural language generation and understanding. These models enable beneficial applications such as content generation, question and answering (Q&A), programming, and code reasoning. Meanwhile, they also pose serious risks by inadvertently or intentionally producing toxic, offensive, or biased content. This dual role of LLMs, both as powerful tools for solving real-world problems and as potential sources of harmful language, presents a pressing sociotechnical challe"},"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":"2508.05775","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-08-07T18:42:16Z","cross_cats_sorted":["cs.CY"],"title_canon_sha256":"b2e7e305791732277c285ba152b7ee9d9cf3d6cae0ea7952aca44062728615ac","abstract_canon_sha256":"35bc93a96ed9e8e53e32f7fb85ed1eae9d77950b5644f6a10a583accfa69a031"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:53:07.015683Z","signature_b64":"9yJOYV1plm7dAzrSaK1OV+w55E6iMzBeAOxYuNAdwhMOSS3CREkAtPbx/gB5FrHXmPcHaVdja4gcMbEU7FqKBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1b78b87700757765d2f6d3c19e30e1ee5bc79a89c5df8b644f9331f207a34512","last_reissued_at":"2026-07-05T11:53:07.015210Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:53:07.015210Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Guardians and Offenders: A Survey on Harmful Content Generation and Safety Mitigation of LLM","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CY"],"primary_cat":"cs.CL","authors_text":"Changjia Zhu, Chi Zhang, Junjie Xiong, Lingyao Li, Xiaoran Xu, Yao Liu, Zhuo Lu","submitted_at":"2025-08-07T18:42:16Z","abstract_excerpt":"Large Language Models (LLMs) have revolutionized content creation across digital platforms, offering unprecedented capabilities in natural language generation and understanding. These models enable beneficial applications such as content generation, question and answering (Q&A), programming, and code reasoning. Meanwhile, they also pose serious risks by inadvertently or intentionally producing toxic, offensive, or biased content. This dual role of LLMs, both as powerful tools for solving real-world problems and as potential sources of harmful language, presents a pressing sociotechnical challe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.05775","kind":"arxiv","version":2},"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/2508.05775/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":"2508.05775","created_at":"2026-07-05T11:53:07.015270+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.05775v2","created_at":"2026-07-05T11:53:07.015270+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.05775","created_at":"2026-07-05T11:53:07.015270+00:00"},{"alias_kind":"pith_short_12","alias_value":"DN4LQ5YAOV3W","created_at":"2026-07-05T11:53:07.015270+00:00"},{"alias_kind":"pith_short_16","alias_value":"DN4LQ5YAOV3WLUXW","created_at":"2026-07-05T11:53:07.015270+00:00"},{"alias_kind":"pith_short_8","alias_value":"DN4LQ5YA","created_at":"2026-07-05T11:53:07.015270+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":5,"sample":[{"citing_arxiv_id":"2602.20102","citing_title":"BarrierSteer: LLM Safety via Learning Barrier Steering","ref_index":24,"is_internal_anchor":true},{"citing_arxiv_id":"2605.14859","citing_title":"Do Coding Agents Understand Least-Privilege Authorization?","ref_index":53,"is_internal_anchor":true},{"citing_arxiv_id":"2605.05630","citing_title":"One Turn Too Late: Response-Aware Defense Against Hidden Malicious Intent in Multi-Turn Dialogue","ref_index":39,"is_internal_anchor":true},{"citing_arxiv_id":"2605.05630","citing_title":"One Turn Too Late: Response-Aware Defense Against Hidden Malicious Intent in Multi-Turn Dialogue","ref_index":39,"is_internal_anchor":true},{"citing_arxiv_id":"2604.11663","citing_title":"Why Do Large Language Models Generate Harmful Content?","ref_index":24,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DN4LQ5YAOV3WLUXW2PAZ4MHB5Z","json":"https://pith.science/pith/DN4LQ5YAOV3WLUXW2PAZ4MHB5Z.json","graph_json":"https://pith.science/api/pith-number/DN4LQ5YAOV3WLUXW2PAZ4MHB5Z/graph.json","events_json":"https://pith.science/api/pith-number/DN4LQ5YAOV3WLUXW2PAZ4MHB5Z/events.json","paper":"https://pith.science/paper/DN4LQ5YA"},"agent_actions":{"view_html":"https://pith.science/pith/DN4LQ5YAOV3WLUXW2PAZ4MHB5Z","download_json":"https://pith.science/pith/DN4LQ5YAOV3WLUXW2PAZ4MHB5Z.json","view_paper":"https://pith.science/paper/DN4LQ5YA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.05775&json=true","fetch_graph":"https://pith.science/api/pith-number/DN4LQ5YAOV3WLUXW2PAZ4MHB5Z/graph.json","fetch_events":"https://pith.science/api/pith-number/DN4LQ5YAOV3WLUXW2PAZ4MHB5Z/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DN4LQ5YAOV3WLUXW2PAZ4MHB5Z/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DN4LQ5YAOV3WLUXW2PAZ4MHB5Z/action/storage_attestation","attest_author":"https://pith.science/pith/DN4LQ5YAOV3WLUXW2PAZ4MHB5Z/action/author_attestation","sign_citation":"https://pith.science/pith/DN4LQ5YAOV3WLUXW2PAZ4MHB5Z/action/citation_signature","submit_replication":"https://pith.science/pith/DN4LQ5YAOV3WLUXW2PAZ4MHB5Z/action/replication_record"}},"created_at":"2026-07-05T11:53:07.015270+00:00","updated_at":"2026-07-05T11:53:07.015270+00:00"}