{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:KMOWH25PVCCYZAW2HANYUYYVGH","short_pith_number":"pith:KMOWH25P","schema_version":"1.0","canonical_sha256":"531d63ebafa8858c82da381b8a631531f4eb7eddd56aaf1f1f687c93e5222155","source":{"kind":"arxiv","id":"2410.03869","version":2},"attestation_state":"computed","paper":{"title":"Chain-of-Jailbreak Attack for Image Generation Models via Editing Step by Step","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CR","cs.CV","cs.MM"],"primary_cat":"cs.CL","authors_text":"Jen-tse Huang, Kuiyi Gao, Qiuzhi Liu, Shuai Wang, Wenxiang Jiao, Wenxuan Wang, Youliang Yuan, Zhaopeng Tu","submitted_at":"2024-10-04T19:04:43Z","abstract_excerpt":"Text-based image generation models, such as Stable Diffusion and DALL-E 3, hold significant potential in content creation and publishing workflows, making them the focus in recent years. Despite their remarkable capability to generate diverse and vivid images, considerable efforts are being made to prevent the generation of harmful content, such as abusive, violent, or pornographic material. To assess the safety of existing models, we introduce a novel jailbreaking method called Chain-of-Jailbreak (CoJ) attack, which compromises image generation models through a step-by-step editing process. S"},"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":"2410.03869","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-04T19:04:43Z","cross_cats_sorted":["cs.AI","cs.CR","cs.CV","cs.MM"],"title_canon_sha256":"4ce4f350715e58a5ada8167bf5ed509ee71ca99d6c0b300c8528a64f6378466a","abstract_canon_sha256":"12458f5a5d2f33350d987f6fa070b7b8447fd241e958998ca35f2423c2d7c251"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:14:49.572851Z","signature_b64":"gq4DSBwCH0L6xce4lMnpw8cgw4CUgtNSZgZCOxeT4/nQ0Q/lSRHbKP1YwIn8b1P/UeSkyhLdpeqNug6zFc5QBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"531d63ebafa8858c82da381b8a631531f4eb7eddd56aaf1f1f687c93e5222155","last_reissued_at":"2026-07-05T11:14:49.572454Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:14:49.572454Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Chain-of-Jailbreak Attack for Image Generation Models via Editing Step by Step","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CR","cs.CV","cs.MM"],"primary_cat":"cs.CL","authors_text":"Jen-tse Huang, Kuiyi Gao, Qiuzhi Liu, Shuai Wang, Wenxiang Jiao, Wenxuan Wang, Youliang Yuan, Zhaopeng Tu","submitted_at":"2024-10-04T19:04:43Z","abstract_excerpt":"Text-based image generation models, such as Stable Diffusion and DALL-E 3, hold significant potential in content creation and publishing workflows, making them the focus in recent years. Despite their remarkable capability to generate diverse and vivid images, considerable efforts are being made to prevent the generation of harmful content, such as abusive, violent, or pornographic material. To assess the safety of existing models, we introduce a novel jailbreaking method called Chain-of-Jailbreak (CoJ) attack, which compromises image generation models through a step-by-step editing process. S"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.03869","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/2410.03869/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":"2410.03869","created_at":"2026-07-05T11:14:49.572507+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.03869v2","created_at":"2026-07-05T11:14:49.572507+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.03869","created_at":"2026-07-05T11:14:49.572507+00:00"},{"alias_kind":"pith_short_12","alias_value":"KMOWH25PVCCY","created_at":"2026-07-05T11:14:49.572507+00:00"},{"alias_kind":"pith_short_16","alias_value":"KMOWH25PVCCYZAW2","created_at":"2026-07-05T11:14:49.572507+00:00"},{"alias_kind":"pith_short_8","alias_value":"KMOWH25P","created_at":"2026-07-05T11:14:49.572507+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.04673","citing_title":"Trojan Horse Prompting: Jailbreaking Conversational Multimodal Models by Forging Assistant Message","ref_index":17,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KMOWH25PVCCYZAW2HANYUYYVGH","json":"https://pith.science/pith/KMOWH25PVCCYZAW2HANYUYYVGH.json","graph_json":"https://pith.science/api/pith-number/KMOWH25PVCCYZAW2HANYUYYVGH/graph.json","events_json":"https://pith.science/api/pith-number/KMOWH25PVCCYZAW2HANYUYYVGH/events.json","paper":"https://pith.science/paper/KMOWH25P"},"agent_actions":{"view_html":"https://pith.science/pith/KMOWH25PVCCYZAW2HANYUYYVGH","download_json":"https://pith.science/pith/KMOWH25PVCCYZAW2HANYUYYVGH.json","view_paper":"https://pith.science/paper/KMOWH25P","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.03869&json=true","fetch_graph":"https://pith.science/api/pith-number/KMOWH25PVCCYZAW2HANYUYYVGH/graph.json","fetch_events":"https://pith.science/api/pith-number/KMOWH25PVCCYZAW2HANYUYYVGH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KMOWH25PVCCYZAW2HANYUYYVGH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KMOWH25PVCCYZAW2HANYUYYVGH/action/storage_attestation","attest_author":"https://pith.science/pith/KMOWH25PVCCYZAW2HANYUYYVGH/action/author_attestation","sign_citation":"https://pith.science/pith/KMOWH25PVCCYZAW2HANYUYYVGH/action/citation_signature","submit_replication":"https://pith.science/pith/KMOWH25PVCCYZAW2HANYUYYVGH/action/replication_record"}},"created_at":"2026-07-05T11:14:49.572507+00:00","updated_at":"2026-07-05T11:14:49.572507+00:00"}