{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:33EWEE64LLPUHJ3PIPUF77ODEM","short_pith_number":"pith:33EWEE64","canonical_record":{"source":{"id":"2403.09792","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-03-14T18:24:55Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"9d7414148a546713f8e3766360d0960e24a8ef17b63ddf3fbe5ecfd0f29c9a43","abstract_canon_sha256":"e3cc4e45bd2717bb8756f986f809d0292cf39330ccbe160eb0a61f6cc14b7209"},"schema_version":"1.0"},"canonical_sha256":"dec96213dc5adf43a76f43e85ffdc32327c40011ba1be9c70c9caa3729423a00","source":{"kind":"arxiv","id":"2403.09792","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.09792","created_at":"2026-07-05T09:59:44Z"},{"alias_kind":"arxiv_version","alias_value":"2403.09792v3","created_at":"2026-07-05T09:59:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.09792","created_at":"2026-07-05T09:59:44Z"},{"alias_kind":"pith_short_12","alias_value":"33EWEE64LLPU","created_at":"2026-07-05T09:59:44Z"},{"alias_kind":"pith_short_16","alias_value":"33EWEE64LLPUHJ3P","created_at":"2026-07-05T09:59:44Z"},{"alias_kind":"pith_short_8","alias_value":"33EWEE64","created_at":"2026-07-05T09:59:44Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:33EWEE64LLPUHJ3PIPUF77ODEM","target":"record","payload":{"canonical_record":{"source":{"id":"2403.09792","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-03-14T18:24:55Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"9d7414148a546713f8e3766360d0960e24a8ef17b63ddf3fbe5ecfd0f29c9a43","abstract_canon_sha256":"e3cc4e45bd2717bb8756f986f809d0292cf39330ccbe160eb0a61f6cc14b7209"},"schema_version":"1.0"},"canonical_sha256":"dec96213dc5adf43a76f43e85ffdc32327c40011ba1be9c70c9caa3729423a00","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:59:44.862405Z","signature_b64":"1J6dOc0P3PBiIVIb7K7n1CFttNImWFe8OAGf9iRFrASFhBFmgs83LQYYQJHOn21ojazzmBd9cLUyUIcpU/KvDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"dec96213dc5adf43a76f43e85ffdc32327c40011ba1be9c70c9caa3729423a00","last_reissued_at":"2026-07-05T09:59:44.861908Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:59:44.861908Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2403.09792","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:59:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DZ+Ced08KEqeSnjNEzVQt1ZREi02ySNnZooh2DHESSUaBEQR1flUI14ydAYMXleH7HBTRO/P7Un4SzlUPjg7DA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T16:48:01.344613Z"},"content_sha256":"2e6716faa8f1637e60b23fcc04d4739d01367d4d718e66ef649fba3a637de31c","schema_version":"1.0","event_id":"sha256:2e6716faa8f1637e60b23fcc04d4739d01367d4d718e66ef649fba3a637de31c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:33EWEE64LLPUHJ3PIPUF77ODEM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Images are Achilles' Heel of Alignment: Exploiting Visual Vulnerabilities for Jailbreaking Multimodal Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Hangyu Guo, Ji-Rong Wen, Kun Zhou, Wayne Xin Zhao, Yifan Li","submitted_at":"2024-03-14T18:24:55Z","abstract_excerpt":"In this paper, we study the harmlessness alignment problem of multimodal large language models (MLLMs). We conduct a systematic empirical analysis of the harmlessness performance of representative MLLMs and reveal that the image input poses the alignment vulnerability of MLLMs. Inspired by this, we propose a novel jailbreak method named HADES, which hides and amplifies the harmfulness of the malicious intent within the text input, using meticulously crafted images. Experimental results show that HADES can effectively jailbreak existing MLLMs, which achieves an average Attack Success Rate (ASR)"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.09792","kind":"arxiv","version":3},"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/2403.09792/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":"2026-07-05T09:59:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"j+lRrzr0gOJePPmckm26KaVjAqdMw8RkPxwDUTNv8rliQ37m+D1aas231j+TH2PRsToVpEcJsNCRWkXYLPxzAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T16:48:01.345165Z"},"content_sha256":"3b5e382fa9fd829fb3b98ee0950d064ffd6c134ffc3ed7cc9de0fc65161c87bc","schema_version":"1.0","event_id":"sha256:3b5e382fa9fd829fb3b98ee0950d064ffd6c134ffc3ed7cc9de0fc65161c87bc"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/33EWEE64LLPUHJ3PIPUF77ODEM/bundle.json","state_url":"https://pith.science/pith/33EWEE64LLPUHJ3PIPUF77ODEM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/33EWEE64LLPUHJ3PIPUF77ODEM/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":"2026-08-07T16:48:01Z","links":{"resolver":"https://pith.science/pith/33EWEE64LLPUHJ3PIPUF77ODEM","bundle":"https://pith.science/pith/33EWEE64LLPUHJ3PIPUF77ODEM/bundle.json","state":"https://pith.science/pith/33EWEE64LLPUHJ3PIPUF77ODEM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/33EWEE64LLPUHJ3PIPUF77ODEM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:33EWEE64LLPUHJ3PIPUF77ODEM","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":"e3cc4e45bd2717bb8756f986f809d0292cf39330ccbe160eb0a61f6cc14b7209","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-03-14T18:24:55Z","title_canon_sha256":"9d7414148a546713f8e3766360d0960e24a8ef17b63ddf3fbe5ecfd0f29c9a43"},"schema_version":"1.0","source":{"id":"2403.09792","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.09792","created_at":"2026-07-05T09:59:44Z"},{"alias_kind":"arxiv_version","alias_value":"2403.09792v3","created_at":"2026-07-05T09:59:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.09792","created_at":"2026-07-05T09:59:44Z"},{"alias_kind":"pith_short_12","alias_value":"33EWEE64LLPU","created_at":"2026-07-05T09:59:44Z"},{"alias_kind":"pith_short_16","alias_value":"33EWEE64LLPUHJ3P","created_at":"2026-07-05T09:59:44Z"},{"alias_kind":"pith_short_8","alias_value":"33EWEE64","created_at":"2026-07-05T09:59:44Z"}],"graph_snapshots":[{"event_id":"sha256:3b5e382fa9fd829fb3b98ee0950d064ffd6c134ffc3ed7cc9de0fc65161c87bc","target":"graph","created_at":"2026-07-05T09:59:44Z","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/2403.09792/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"In this paper, we study the harmlessness alignment problem of multimodal large language models (MLLMs). We conduct a systematic empirical analysis of the harmlessness performance of representative MLLMs and reveal that the image input poses the alignment vulnerability of MLLMs. Inspired by this, we propose a novel jailbreak method named HADES, which hides and amplifies the harmfulness of the malicious intent within the text input, using meticulously crafted images. Experimental results show that HADES can effectively jailbreak existing MLLMs, which achieves an average Attack Success Rate (ASR)","authors_text":"Hangyu Guo, Ji-Rong Wen, Kun Zhou, Wayne Xin Zhao, Yifan Li","cross_cats":["cs.CL"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-03-14T18:24:55Z","title":"Images are Achilles' Heel of Alignment: Exploiting Visual Vulnerabilities for Jailbreaking Multimodal Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.09792","kind":"arxiv","version":3},"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":"sha256:2e6716faa8f1637e60b23fcc04d4739d01367d4d718e66ef649fba3a637de31c","target":"record","created_at":"2026-07-05T09:59:44Z","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":"e3cc4e45bd2717bb8756f986f809d0292cf39330ccbe160eb0a61f6cc14b7209","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-03-14T18:24:55Z","title_canon_sha256":"9d7414148a546713f8e3766360d0960e24a8ef17b63ddf3fbe5ecfd0f29c9a43"},"schema_version":"1.0","source":{"id":"2403.09792","kind":"arxiv","version":3}},"canonical_sha256":"dec96213dc5adf43a76f43e85ffdc32327c40011ba1be9c70c9caa3729423a00","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"dec96213dc5adf43a76f43e85ffdc32327c40011ba1be9c70c9caa3729423a00","first_computed_at":"2026-07-05T09:59:44.861908Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:59:44.861908Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"1J6dOc0P3PBiIVIb7K7n1CFttNImWFe8OAGf9iRFrASFhBFmgs83LQYYQJHOn21ojazzmBd9cLUyUIcpU/KvDg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:59:44.862405Z","signed_message":"canonical_sha256_bytes"},"source_id":"2403.09792","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2e6716faa8f1637e60b23fcc04d4739d01367d4d718e66ef649fba3a637de31c","sha256:3b5e382fa9fd829fb3b98ee0950d064ffd6c134ffc3ed7cc9de0fc65161c87bc"],"state_sha256":"05679d6058c7e0b9397b64aba0bbd55b41f637939d384b408ca98fff0aabd4bb"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"alYMhCeOgpX5aMJ5Hxkp+444lpkJhILAtf1ZWRIa3hm9QrOD5OLwNPj8d/n69QYl1TZV7I1sOLWsdqUSzIx7BQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T16:48:01.349343Z","bundle_sha256":"5847d0a145c5fe4947c8ab8815e766b9cce5c7f11e03846625e82a34e436ffc8"}}