{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:Z6WKJI5SFETOQWYIZWIGLCABA2","short_pith_number":"pith:Z6WKJI5S","canonical_record":{"source":{"id":"2410.18927","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-10-24T17:14:40Z","cross_cats_sorted":[],"title_canon_sha256":"5ba11b74c9b4ea23f557373cd2c61a48ae8601cad16e38034120f3d944adbacc","abstract_canon_sha256":"8d90efe6eb03b003b758c59f76146498f477c88e0222d52d9548a065bdee6bf2"},"schema_version":"1.0"},"canonical_sha256":"cfaca4a3b22926e85b08cd9065880106a8a2392230c5442fd26da858b3d1e93c","source":{"kind":"arxiv","id":"2410.18927","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.18927","created_at":"2026-07-05T09:25:22Z"},{"alias_kind":"arxiv_version","alias_value":"2410.18927v1","created_at":"2026-07-05T09:25:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.18927","created_at":"2026-07-05T09:25:22Z"},{"alias_kind":"pith_short_12","alias_value":"Z6WKJI5SFETO","created_at":"2026-07-05T09:25:22Z"},{"alias_kind":"pith_short_16","alias_value":"Z6WKJI5SFETOQWYI","created_at":"2026-07-05T09:25:22Z"},{"alias_kind":"pith_short_8","alias_value":"Z6WKJI5S","created_at":"2026-07-05T09:25:22Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:Z6WKJI5SFETOQWYIZWIGLCABA2","target":"record","payload":{"canonical_record":{"source":{"id":"2410.18927","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-10-24T17:14:40Z","cross_cats_sorted":[],"title_canon_sha256":"5ba11b74c9b4ea23f557373cd2c61a48ae8601cad16e38034120f3d944adbacc","abstract_canon_sha256":"8d90efe6eb03b003b758c59f76146498f477c88e0222d52d9548a065bdee6bf2"},"schema_version":"1.0"},"canonical_sha256":"cfaca4a3b22926e85b08cd9065880106a8a2392230c5442fd26da858b3d1e93c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:25:22.122858Z","signature_b64":"z476uWAqHsKzE583RdevfDiiFZ9S/fpgrsIILcOmrLU8t6dC7P/uBxDVV90fovrAjWtVbfaCQL+oIqzJaQ7yAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cfaca4a3b22926e85b08cd9065880106a8a2392230c5442fd26da858b3d1e93c","last_reissued_at":"2026-07-05T09:25:22.122350Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:25:22.122350Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2410.18927","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:25:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wEy9Dz0xTMaLwVxcJSuT/9OqJUviMNQXqqmjZ9pu22d8S9iKeG5wg/OVBp+m16m1QXvDbkcQxtZyRPQvBNp/Dw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T11:59:23.001158Z"},"content_sha256":"94581a5d1de1619baf4cc38dccac348dae64c15d53062c8f655206f84f2ee344","schema_version":"1.0","event_id":"sha256:94581a5d1de1619baf4cc38dccac348dae64c15d53062c8f655206f84f2ee344"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:Z6WKJI5SFETOQWYIZWIGLCABA2","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"SafeBench: A Safety Evaluation Framework for Multimodal Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CR","authors_text":"Aishan Liu, Dacheng Tao, Jinyang Guo, Lei Huang, Siyuan Liang, Wenbo Zhou, Xianglong Liu, Zonghao Ying","submitted_at":"2024-10-24T17:14:40Z","abstract_excerpt":"Multimodal Large Language Models (MLLMs) are showing strong safety concerns (e.g., generating harmful outputs for users), which motivates the development of safety evaluation benchmarks. However, we observe that existing safety benchmarks for MLLMs show limitations in query quality and evaluation reliability limiting the detection of model safety implications as MLLMs continue to evolve. In this paper, we propose \\toolns, a comprehensive framework designed for conducting safety evaluations of MLLMs. Our framework consists of a comprehensive harmful query dataset and an automated evaluation pro"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.18927","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/2410.18927/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":"2026-07-05T09:25:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"NF2KF3Md1YkQI4U5g3sKL23vufVmRXWGzqU+6CUZFENKeXu0UjoANNBgg765cVkNiEl97rArRnwMMWWvyrosBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T11:59:23.001678Z"},"content_sha256":"1101f8c6e69d0a17c64744877d677b0b49bde66260c8fb7d4b4eeb4ea5352dfe","schema_version":"1.0","event_id":"sha256:1101f8c6e69d0a17c64744877d677b0b49bde66260c8fb7d4b4eeb4ea5352dfe"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/Z6WKJI5SFETOQWYIZWIGLCABA2/bundle.json","state_url":"https://pith.science/pith/Z6WKJI5SFETOQWYIZWIGLCABA2/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/Z6WKJI5SFETOQWYIZWIGLCABA2/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":"2026-08-04T11:59:23Z","links":{"resolver":"https://pith.science/pith/Z6WKJI5SFETOQWYIZWIGLCABA2","bundle":"https://pith.science/pith/Z6WKJI5SFETOQWYIZWIGLCABA2/bundle.json","state":"https://pith.science/pith/Z6WKJI5SFETOQWYIZWIGLCABA2/state.json","well_known_bundle":"https://pith.science/.well-known/pith/Z6WKJI5SFETOQWYIZWIGLCABA2/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:Z6WKJI5SFETOQWYIZWIGLCABA2","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":"8d90efe6eb03b003b758c59f76146498f477c88e0222d52d9548a065bdee6bf2","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-10-24T17:14:40Z","title_canon_sha256":"5ba11b74c9b4ea23f557373cd2c61a48ae8601cad16e38034120f3d944adbacc"},"schema_version":"1.0","source":{"id":"2410.18927","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.18927","created_at":"2026-07-05T09:25:22Z"},{"alias_kind":"arxiv_version","alias_value":"2410.18927v1","created_at":"2026-07-05T09:25:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.18927","created_at":"2026-07-05T09:25:22Z"},{"alias_kind":"pith_short_12","alias_value":"Z6WKJI5SFETO","created_at":"2026-07-05T09:25:22Z"},{"alias_kind":"pith_short_16","alias_value":"Z6WKJI5SFETOQWYI","created_at":"2026-07-05T09:25:22Z"},{"alias_kind":"pith_short_8","alias_value":"Z6WKJI5S","created_at":"2026-07-05T09:25:22Z"}],"graph_snapshots":[{"event_id":"sha256:1101f8c6e69d0a17c64744877d677b0b49bde66260c8fb7d4b4eeb4ea5352dfe","target":"graph","created_at":"2026-07-05T09:25:22Z","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/2410.18927/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Multimodal Large Language Models (MLLMs) are showing strong safety concerns (e.g., generating harmful outputs for users), which motivates the development of safety evaluation benchmarks. However, we observe that existing safety benchmarks for MLLMs show limitations in query quality and evaluation reliability limiting the detection of model safety implications as MLLMs continue to evolve. In this paper, we propose \\toolns, a comprehensive framework designed for conducting safety evaluations of MLLMs. Our framework consists of a comprehensive harmful query dataset and an automated evaluation pro","authors_text":"Aishan Liu, Dacheng Tao, Jinyang Guo, Lei Huang, Siyuan Liang, Wenbo Zhou, Xianglong Liu, Zonghao Ying","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-10-24T17:14:40Z","title":"SafeBench: A Safety Evaluation Framework for Multimodal Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.18927","kind":"arxiv","version":1},"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":"sha256:94581a5d1de1619baf4cc38dccac348dae64c15d53062c8f655206f84f2ee344","target":"record","created_at":"2026-07-05T09:25:22Z","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":"8d90efe6eb03b003b758c59f76146498f477c88e0222d52d9548a065bdee6bf2","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-10-24T17:14:40Z","title_canon_sha256":"5ba11b74c9b4ea23f557373cd2c61a48ae8601cad16e38034120f3d944adbacc"},"schema_version":"1.0","source":{"id":"2410.18927","kind":"arxiv","version":1}},"canonical_sha256":"cfaca4a3b22926e85b08cd9065880106a8a2392230c5442fd26da858b3d1e93c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"cfaca4a3b22926e85b08cd9065880106a8a2392230c5442fd26da858b3d1e93c","first_computed_at":"2026-07-05T09:25:22.122350Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:25:22.122350Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"z476uWAqHsKzE583RdevfDiiFZ9S/fpgrsIILcOmrLU8t6dC7P/uBxDVV90fovrAjWtVbfaCQL+oIqzJaQ7yAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:25:22.122858Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.18927","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:94581a5d1de1619baf4cc38dccac348dae64c15d53062c8f655206f84f2ee344","sha256:1101f8c6e69d0a17c64744877d677b0b49bde66260c8fb7d4b4eeb4ea5352dfe"],"state_sha256":"51946ef5291aebcdc5c98c0d31addc172f0b47fef993b0292e51edb671180d5f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VgcAaHj8sziBOuPuSQ60pzIH89swMpOPRZhs17+eAkjXRKzvHDgovHguxi9KWaZKNVT/5ORntHR+TNPQcdV2Cw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T11:59:23.005568Z","bundle_sha256":"8bb559eddcf532c0981406087318cbc69d5acbf794fc1676efd75592d9f32614"}}