{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:O2QSQXCVJ77BADNF7MNWH2IFH6","short_pith_number":"pith:O2QSQXCV","schema_version":"1.0","canonical_sha256":"76a1285c554ffe100da5fb1b63e9053fb8dd68d50df4794f46d101e77c510c48","source":{"kind":"arxiv","id":"2412.10423","version":2},"attestation_state":"computed","paper":{"title":"Look Before You Leap: Enhancing Attention and Vigilance Regarding Harmful Content with GuidelineLLM","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Kehai Chen, Min Zhang, Muyun Yang, Rongxiang Weng, Shaoqing Zhang, Tiejun Zhao, Zhuosheng Zhang","submitted_at":"2024-12-10T12:42:33Z","abstract_excerpt":"Despite being empowered with alignment mechanisms, large language models (LLMs) are increasingly vulnerable to emerging jailbreak attacks that can compromise their alignment mechanisms. This vulnerability poses significant risks to real-world applications. Existing work faces challenges in both training efficiency and generalization capabilities (i.e., Reinforcement Learning from Human Feedback and Red-Teaming). Developing effective strategies to enable LLMs to resist continuously evolving jailbreak attempts represents a significant challenge. To address this challenge, we propose a novel defe"},"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":"2412.10423","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-12-10T12:42:33Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3c51cf47fd270077ae75fc40d9856578131e791c9ad24176e068c396d033ced4","abstract_canon_sha256":"eb0e330078263ca04a27d8cced2864a8d8fb127702b48ca8edb1040c2e937dbe"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:48:22.699832Z","signature_b64":"kR+4WStSHimTZDlXSuNw3ZZPbDSCpRXLXiZhnNwVn01xvYnqVfR03EdisohD2y9dA7gCvidR3cMz+2gizD6FDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"76a1285c554ffe100da5fb1b63e9053fb8dd68d50df4794f46d101e77c510c48","last_reissued_at":"2026-07-05T10:48:22.699385Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:48:22.699385Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Look Before You Leap: Enhancing Attention and Vigilance Regarding Harmful Content with GuidelineLLM","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Kehai Chen, Min Zhang, Muyun Yang, Rongxiang Weng, Shaoqing Zhang, Tiejun Zhao, Zhuosheng Zhang","submitted_at":"2024-12-10T12:42:33Z","abstract_excerpt":"Despite being empowered with alignment mechanisms, large language models (LLMs) are increasingly vulnerable to emerging jailbreak attacks that can compromise their alignment mechanisms. This vulnerability poses significant risks to real-world applications. Existing work faces challenges in both training efficiency and generalization capabilities (i.e., Reinforcement Learning from Human Feedback and Red-Teaming). Developing effective strategies to enable LLMs to resist continuously evolving jailbreak attempts represents a significant challenge. To address this challenge, we propose a novel defe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.10423","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/2412.10423/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":"2412.10423","created_at":"2026-07-05T10:48:22.699442+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.10423v2","created_at":"2026-07-05T10:48:22.699442+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.10423","created_at":"2026-07-05T10:48:22.699442+00:00"},{"alias_kind":"pith_short_12","alias_value":"O2QSQXCVJ77B","created_at":"2026-07-05T10:48:22.699442+00:00"},{"alias_kind":"pith_short_16","alias_value":"O2QSQXCVJ77BADNF","created_at":"2026-07-05T10:48:22.699442+00:00"},{"alias_kind":"pith_short_8","alias_value":"O2QSQXCV","created_at":"2026-07-05T10:48:22.699442+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/O2QSQXCVJ77BADNF7MNWH2IFH6","json":"https://pith.science/pith/O2QSQXCVJ77BADNF7MNWH2IFH6.json","graph_json":"https://pith.science/api/pith-number/O2QSQXCVJ77BADNF7MNWH2IFH6/graph.json","events_json":"https://pith.science/api/pith-number/O2QSQXCVJ77BADNF7MNWH2IFH6/events.json","paper":"https://pith.science/paper/O2QSQXCV"},"agent_actions":{"view_html":"https://pith.science/pith/O2QSQXCVJ77BADNF7MNWH2IFH6","download_json":"https://pith.science/pith/O2QSQXCVJ77BADNF7MNWH2IFH6.json","view_paper":"https://pith.science/paper/O2QSQXCV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.10423&json=true","fetch_graph":"https://pith.science/api/pith-number/O2QSQXCVJ77BADNF7MNWH2IFH6/graph.json","fetch_events":"https://pith.science/api/pith-number/O2QSQXCVJ77BADNF7MNWH2IFH6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/O2QSQXCVJ77BADNF7MNWH2IFH6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/O2QSQXCVJ77BADNF7MNWH2IFH6/action/storage_attestation","attest_author":"https://pith.science/pith/O2QSQXCVJ77BADNF7MNWH2IFH6/action/author_attestation","sign_citation":"https://pith.science/pith/O2QSQXCVJ77BADNF7MNWH2IFH6/action/citation_signature","submit_replication":"https://pith.science/pith/O2QSQXCVJ77BADNF7MNWH2IFH6/action/replication_record"}},"created_at":"2026-07-05T10:48:22.699442+00:00","updated_at":"2026-07-05T10:48:22.699442+00:00"}