{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:ZLFVPUN366MJTO765NJXMWN6OK","short_pith_number":"pith:ZLFVPUN3","schema_version":"1.0","canonical_sha256":"cacb57d1bbf79899bbfeeb537659be72b63f815f9f840b9ac167748ee73b1daa","source":{"kind":"arxiv","id":"2406.07057","version":2},"attestation_state":"computed","paper":{"title":"MultiTrust: A Comprehensive Benchmark Towards Trustworthy Multimodal Large Language Models","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.LG"],"primary_cat":"cs.CL","authors_text":"Chang Liu, Hang Su, Huanran Chen, Jun Zhu, Xiao Yang, Xingxing Wei, Yao Huang, Yichi Zhang, Yifan Wang, Yinpeng Dong, Yitong Sun, Zhengwei Fang, Zhe Zhao","submitted_at":"2024-06-11T08:38:13Z","abstract_excerpt":"Despite the superior capabilities of Multimodal Large Language Models (MLLMs) across diverse tasks, they still face significant trustworthiness challenges. Yet, current literature on the assessment of trustworthy MLLMs remains limited, lacking a holistic evaluation to offer thorough insights into future improvements. In this work, we establish MultiTrust, the first comprehensive and unified benchmark on the trustworthiness of MLLMs across five primary aspects: truthfulness, safety, robustness, fairness, and privacy. Our benchmark employs a rigorous evaluation strategy that addresses both multi"},"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":"2406.07057","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-11T08:38:13Z","cross_cats_sorted":["cs.AI","cs.CV","cs.LG"],"title_canon_sha256":"31f1dabbfccf598e75a266a522af835c1973558a83ca235089805d691892f01c","abstract_canon_sha256":"ae8bce9df905d7a9a4546ac438d09e6cfca08d32e3ca5d952107edcb39892047"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:45:10.857604Z","signature_b64":"aL+8fxeIJ4y9iwWBZwb1ugMZmMJffaHBlX1sUD4uH382ua/fuKx4zyno8dB9V0r2fQjCXCwEY/kuD+chngMqCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cacb57d1bbf79899bbfeeb537659be72b63f815f9f840b9ac167748ee73b1daa","last_reissued_at":"2026-07-05T09:45:10.857112Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:45:10.857112Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MultiTrust: A Comprehensive Benchmark Towards Trustworthy Multimodal Large Language Models","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.LG"],"primary_cat":"cs.CL","authors_text":"Chang Liu, Hang Su, Huanran Chen, Jun Zhu, Xiao Yang, Xingxing Wei, Yao Huang, Yichi Zhang, Yifan Wang, Yinpeng Dong, Yitong Sun, Zhengwei Fang, Zhe Zhao","submitted_at":"2024-06-11T08:38:13Z","abstract_excerpt":"Despite the superior capabilities of Multimodal Large Language Models (MLLMs) across diverse tasks, they still face significant trustworthiness challenges. Yet, current literature on the assessment of trustworthy MLLMs remains limited, lacking a holistic evaluation to offer thorough insights into future improvements. In this work, we establish MultiTrust, the first comprehensive and unified benchmark on the trustworthiness of MLLMs across five primary aspects: truthfulness, safety, robustness, fairness, and privacy. Our benchmark employs a rigorous evaluation strategy that addresses both multi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.07057","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/2406.07057/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":"2406.07057","created_at":"2026-07-05T09:45:10.857167+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.07057v2","created_at":"2026-07-05T09:45:10.857167+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.07057","created_at":"2026-07-05T09:45:10.857167+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZLFVPUN366MJ","created_at":"2026-07-05T09:45:10.857167+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZLFVPUN366MJTO76","created_at":"2026-07-05T09:45:10.857167+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZLFVPUN3","created_at":"2026-07-05T09:45:10.857167+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.03713","citing_title":"Investigating Adversarial Robustness of Multi-modal Large Language Models","ref_index":68,"is_internal_anchor":false},{"citing_arxiv_id":"2501.13340","citing_title":"Retrievals Can Be Detrimental: Unveiling the Backdoor Vulnerability of Retrieval-Augmented Diffusion Models","ref_index":57,"is_internal_anchor":false},{"citing_arxiv_id":"2506.07180","citing_title":"Flattery in Motion: Benchmarking and Analyzing Sycophancy in Video-LLMs","ref_index":40,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZLFVPUN366MJTO765NJXMWN6OK","json":"https://pith.science/pith/ZLFVPUN366MJTO765NJXMWN6OK.json","graph_json":"https://pith.science/api/pith-number/ZLFVPUN366MJTO765NJXMWN6OK/graph.json","events_json":"https://pith.science/api/pith-number/ZLFVPUN366MJTO765NJXMWN6OK/events.json","paper":"https://pith.science/paper/ZLFVPUN3"},"agent_actions":{"view_html":"https://pith.science/pith/ZLFVPUN366MJTO765NJXMWN6OK","download_json":"https://pith.science/pith/ZLFVPUN366MJTO765NJXMWN6OK.json","view_paper":"https://pith.science/paper/ZLFVPUN3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.07057&json=true","fetch_graph":"https://pith.science/api/pith-number/ZLFVPUN366MJTO765NJXMWN6OK/graph.json","fetch_events":"https://pith.science/api/pith-number/ZLFVPUN366MJTO765NJXMWN6OK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZLFVPUN366MJTO765NJXMWN6OK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZLFVPUN366MJTO765NJXMWN6OK/action/storage_attestation","attest_author":"https://pith.science/pith/ZLFVPUN366MJTO765NJXMWN6OK/action/author_attestation","sign_citation":"https://pith.science/pith/ZLFVPUN366MJTO765NJXMWN6OK/action/citation_signature","submit_replication":"https://pith.science/pith/ZLFVPUN366MJTO765NJXMWN6OK/action/replication_record"}},"created_at":"2026-07-05T09:45:10.857167+00:00","updated_at":"2026-07-05T09:45:10.857167+00:00"}