{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:HPOTDQSSM66OCPKNDGH23J4O24","short_pith_number":"pith:HPOTDQSS","canonical_record":{"source":{"id":"2503.06989","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-03-10T07:10:38Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"1f60725492e774ca79ec0f0c0d13b02e03e6f42776b1b0116f2e02a4bb7527e8","abstract_canon_sha256":"fe9ea7fc8a1b281c8cedf764e43e5559570d21258061110abaf9cc502b883445"},"schema_version":"1.0"},"canonical_sha256":"3bdd31c25267bce13d4d198fada78ed737124056daa38a8303784c9202eb18ea","source":{"kind":"arxiv","id":"2503.06989","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.06989","created_at":"2026-07-05T12:00:46Z"},{"alias_kind":"arxiv_version","alias_value":"2503.06989v4","created_at":"2026-07-05T12:00:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.06989","created_at":"2026-07-05T12:00:46Z"},{"alias_kind":"pith_short_12","alias_value":"HPOTDQSSM66O","created_at":"2026-07-05T12:00:46Z"},{"alias_kind":"pith_short_16","alias_value":"HPOTDQSSM66OCPKN","created_at":"2026-07-05T12:00:46Z"},{"alias_kind":"pith_short_8","alias_value":"HPOTDQSS","created_at":"2026-07-05T12:00:46Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:HPOTDQSSM66OCPKNDGH23J4O24","target":"record","payload":{"canonical_record":{"source":{"id":"2503.06989","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-03-10T07:10:38Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"1f60725492e774ca79ec0f0c0d13b02e03e6f42776b1b0116f2e02a4bb7527e8","abstract_canon_sha256":"fe9ea7fc8a1b281c8cedf764e43e5559570d21258061110abaf9cc502b883445"},"schema_version":"1.0"},"canonical_sha256":"3bdd31c25267bce13d4d198fada78ed737124056daa38a8303784c9202eb18ea","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:00:46.745249Z","signature_b64":"2glm0MhcP7XuzjBitf3ORZF1KL2Bb51mlJb1JYUv143ONmJweGREVm5Stzhf21e1OaXfsmJluORbXmzCEfBjDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3bdd31c25267bce13d4d198fada78ed737124056daa38a8303784c9202eb18ea","last_reissued_at":"2026-07-05T12:00:46.744740Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:00:46.744740Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2503.06989","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T12:00:46Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kN5m7nTIln9U+RIEzrdNQC1FxNzhFhHrMUz+nWfTippSg4Tgpr96V7Iu429xIOAG0embe8ptbrhn2LILgAS1CA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T02:47:03.971902Z"},"content_sha256":"0c461b4a808297c3211981d22733c1f6ea3706c29d14d17fc13f9a4d7c23bde4","schema_version":"1.0","event_id":"sha256:0c461b4a808297c3211981d22733c1f6ea3706c29d14d17fc13f9a4d7c23bde4"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:HPOTDQSSM66OCPKNDGH23J4O24","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Probabilistic Modeling of Jailbreak on Multimodal LLMs: From Quantification to Application","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.CR","authors_text":"Deyue Zhang, Dongdong Yang, Quanchen Zou, Wenzhuo Xu, Xiangzheng Zhang, Xiongtao Sun, Zhipeng Wei, Zonghao Ying","submitted_at":"2025-03-10T07:10:38Z","abstract_excerpt":"Recently, Multimodal Large Language Models (MLLMs) have demonstrated their superior ability in understanding multimodal content. However, they remain vulnerable to jailbreak attacks, which exploit weaknesses in their safety alignment to generate harmful responses. Previous studies categorize jailbreaks as successful or failed based on whether responses contain malicious content. However, given the stochastic nature of MLLM responses, this binary classification of an input's ability to jailbreak MLLMs is inappropriate. Derived from this viewpoint, we introduce jailbreak probability to quantify "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.06989","kind":"arxiv","version":4},"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/2503.06989/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T12:00:46Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Jgk2ljeNbUj8pOFy5smzGBP0QY4M0i7iegCQubYoOsAaBvPfaxvvsMleUCZHCtfuQAfzZs0eS9aSdXA6Zt5NDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T02:47:03.972804Z"},"content_sha256":"1c7e11f8b7aae089551eae1ca2e2a3b51f0227d8df3d4f3d626fc521a557fd3d","schema_version":"1.0","event_id":"sha256:1c7e11f8b7aae089551eae1ca2e2a3b51f0227d8df3d4f3d626fc521a557fd3d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HPOTDQSSM66OCPKNDGH23J4O24/bundle.json","state_url":"https://pith.science/pith/HPOTDQSSM66OCPKNDGH23J4O24/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HPOTDQSSM66OCPKNDGH23J4O24/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-06T02:47:03Z","links":{"resolver":"https://pith.science/pith/HPOTDQSSM66OCPKNDGH23J4O24","bundle":"https://pith.science/pith/HPOTDQSSM66OCPKNDGH23J4O24/bundle.json","state":"https://pith.science/pith/HPOTDQSSM66OCPKNDGH23J4O24/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HPOTDQSSM66OCPKNDGH23J4O24/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:HPOTDQSSM66OCPKNDGH23J4O24","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"fe9ea7fc8a1b281c8cedf764e43e5559570d21258061110abaf9cc502b883445","cross_cats_sorted":["cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-03-10T07:10:38Z","title_canon_sha256":"1f60725492e774ca79ec0f0c0d13b02e03e6f42776b1b0116f2e02a4bb7527e8"},"schema_version":"1.0","source":{"id":"2503.06989","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.06989","created_at":"2026-07-05T12:00:46Z"},{"alias_kind":"arxiv_version","alias_value":"2503.06989v4","created_at":"2026-07-05T12:00:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.06989","created_at":"2026-07-05T12:00:46Z"},{"alias_kind":"pith_short_12","alias_value":"HPOTDQSSM66O","created_at":"2026-07-05T12:00:46Z"},{"alias_kind":"pith_short_16","alias_value":"HPOTDQSSM66OCPKN","created_at":"2026-07-05T12:00:46Z"},{"alias_kind":"pith_short_8","alias_value":"HPOTDQSS","created_at":"2026-07-05T12:00:46Z"}],"graph_snapshots":[{"event_id":"sha256:1c7e11f8b7aae089551eae1ca2e2a3b51f0227d8df3d4f3d626fc521a557fd3d","target":"graph","created_at":"2026-07-05T12:00:46Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2503.06989/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recently, Multimodal Large Language Models (MLLMs) have demonstrated their superior ability in understanding multimodal content. However, they remain vulnerable to jailbreak attacks, which exploit weaknesses in their safety alignment to generate harmful responses. Previous studies categorize jailbreaks as successful or failed based on whether responses contain malicious content. However, given the stochastic nature of MLLM responses, this binary classification of an input's ability to jailbreak MLLMs is inappropriate. Derived from this viewpoint, we introduce jailbreak probability to quantify ","authors_text":"Deyue Zhang, Dongdong Yang, Quanchen Zou, Wenzhuo Xu, Xiangzheng Zhang, Xiongtao Sun, Zhipeng Wei, Zonghao Ying","cross_cats":["cs.CV"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-03-10T07:10:38Z","title":"Probabilistic Modeling of Jailbreak on Multimodal LLMs: From Quantification to Application"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.06989","kind":"arxiv","version":4},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:0c461b4a808297c3211981d22733c1f6ea3706c29d14d17fc13f9a4d7c23bde4","target":"record","created_at":"2026-07-05T12:00:46Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"fe9ea7fc8a1b281c8cedf764e43e5559570d21258061110abaf9cc502b883445","cross_cats_sorted":["cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-03-10T07:10:38Z","title_canon_sha256":"1f60725492e774ca79ec0f0c0d13b02e03e6f42776b1b0116f2e02a4bb7527e8"},"schema_version":"1.0","source":{"id":"2503.06989","kind":"arxiv","version":4}},"canonical_sha256":"3bdd31c25267bce13d4d198fada78ed737124056daa38a8303784c9202eb18ea","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3bdd31c25267bce13d4d198fada78ed737124056daa38a8303784c9202eb18ea","first_computed_at":"2026-07-05T12:00:46.744740Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T12:00:46.744740Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"2glm0MhcP7XuzjBitf3ORZF1KL2Bb51mlJb1JYUv143ONmJweGREVm5Stzhf21e1OaXfsmJluORbXmzCEfBjDw==","signature_status":"signed_v1","signed_at":"2026-07-05T12:00:46.745249Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.06989","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0c461b4a808297c3211981d22733c1f6ea3706c29d14d17fc13f9a4d7c23bde4","sha256:1c7e11f8b7aae089551eae1ca2e2a3b51f0227d8df3d4f3d626fc521a557fd3d"],"state_sha256":"8da17a3bb099b51888d6e92dd40897719dfc8a102e0c2629a1253a1305569884"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"J17JBKsPKfFxxAJJEAzrCi5JJaDFeIzP04teRZ7YNbUH87RsZiro1h7EaLIG/1tlIGysQq+4wMPxd0pCwaucBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T02:47:03.991642Z","bundle_sha256":"f79a6755a8668dfc7439d04bc63c9656494dd1453504fe62d8dbf20c9e592e1c"}}