{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:HOZMBXDC7O7CCQLHYQLF472BL6","short_pith_number":"pith:HOZMBXDC","schema_version":"1.0","canonical_sha256":"3bb2c0dc62fbbe214167c4165e7f415fb5317b1451e8e0654243c9db7e76b118","source":{"kind":"arxiv","id":"2505.08804","version":1},"attestation_state":"computed","paper":{"title":"TokenProber: Jailbreaking Text-to-image Models via Fine-grained Word Impact Analysis","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CR","authors_text":"Chao Shen, Longtian Wang, Tianlin Li, Xiaofei Xie, Yuhan Zhi","submitted_at":"2025-05-11T06:32:33Z","abstract_excerpt":"Text-to-image (T2I) models have significantly advanced in producing high-quality images. However, such models have the ability to generate images containing not-safe-for-work (NSFW) content, such as pornography, violence, political content, and discrimination. To mitigate the risk of generating NSFW content, refusal mechanisms, i.e., safety checkers, have been developed to check potential NSFW content. Adversarial prompting techniques have been developed to evaluate the robustness of the refusal mechanisms. The key challenge remains to subtly modify the prompt in a way that preserves its sensi"},"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":"2505.08804","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-05-11T06:32:33Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"52b4ff8a7543517da59745819e3cf2fdcc2d78cce13dc135f41154812aba1d17","abstract_canon_sha256":"611d8681fb3d1dd2a8bc373d9780eae722d0ebe89674235cb7c0d447ea098828"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:02:44.630190Z","signature_b64":"nmfdzXu+jiRPlLtDrK7o61QNgh1Drn8gqrTf4pWEVWOSE73+dBfheFcx0ZbEdDaktmRBM50F0yIbH66uWe6MBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3bb2c0dc62fbbe214167c4165e7f415fb5317b1451e8e0654243c9db7e76b118","last_reissued_at":"2026-07-05T11:02:44.629676Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:02:44.629676Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TokenProber: Jailbreaking Text-to-image Models via Fine-grained Word Impact Analysis","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CR","authors_text":"Chao Shen, Longtian Wang, Tianlin Li, Xiaofei Xie, Yuhan Zhi","submitted_at":"2025-05-11T06:32:33Z","abstract_excerpt":"Text-to-image (T2I) models have significantly advanced in producing high-quality images. However, such models have the ability to generate images containing not-safe-for-work (NSFW) content, such as pornography, violence, political content, and discrimination. To mitigate the risk of generating NSFW content, refusal mechanisms, i.e., safety checkers, have been developed to check potential NSFW content. Adversarial prompting techniques have been developed to evaluate the robustness of the refusal mechanisms. The key challenge remains to subtly modify the prompt in a way that preserves its sensi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.08804","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/2505.08804/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":"2505.08804","created_at":"2026-07-05T11:02:44.629735+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.08804v1","created_at":"2026-07-05T11:02:44.629735+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.08804","created_at":"2026-07-05T11:02:44.629735+00:00"},{"alias_kind":"pith_short_12","alias_value":"HOZMBXDC7O7C","created_at":"2026-07-05T11:02:44.629735+00:00"},{"alias_kind":"pith_short_16","alias_value":"HOZMBXDC7O7CCQLH","created_at":"2026-07-05T11:02:44.629735+00:00"},{"alias_kind":"pith_short_8","alias_value":"HOZMBXDC","created_at":"2026-07-05T11:02:44.629735+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/HOZMBXDC7O7CCQLHYQLF472BL6","json":"https://pith.science/pith/HOZMBXDC7O7CCQLHYQLF472BL6.json","graph_json":"https://pith.science/api/pith-number/HOZMBXDC7O7CCQLHYQLF472BL6/graph.json","events_json":"https://pith.science/api/pith-number/HOZMBXDC7O7CCQLHYQLF472BL6/events.json","paper":"https://pith.science/paper/HOZMBXDC"},"agent_actions":{"view_html":"https://pith.science/pith/HOZMBXDC7O7CCQLHYQLF472BL6","download_json":"https://pith.science/pith/HOZMBXDC7O7CCQLHYQLF472BL6.json","view_paper":"https://pith.science/paper/HOZMBXDC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.08804&json=true","fetch_graph":"https://pith.science/api/pith-number/HOZMBXDC7O7CCQLHYQLF472BL6/graph.json","fetch_events":"https://pith.science/api/pith-number/HOZMBXDC7O7CCQLHYQLF472BL6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HOZMBXDC7O7CCQLHYQLF472BL6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HOZMBXDC7O7CCQLHYQLF472BL6/action/storage_attestation","attest_author":"https://pith.science/pith/HOZMBXDC7O7CCQLHYQLF472BL6/action/author_attestation","sign_citation":"https://pith.science/pith/HOZMBXDC7O7CCQLHYQLF472BL6/action/citation_signature","submit_replication":"https://pith.science/pith/HOZMBXDC7O7CCQLHYQLF472BL6/action/replication_record"}},"created_at":"2026-07-05T11:02:44.629735+00:00","updated_at":"2026-07-05T11:02:44.629735+00:00"}