arXiv cs.CL· Junhao Wang, Yidou Chen, Li Lu, Hongwei Yao, Xiaodan Li, Mengjie Wu, Yuefeng Chen, Zhan Qin, Kui Ren·· 12 小时前AI 评分38
Omni-UBench：跨安全、实用性与可靠性的全模态 LLM 护栏基准测试
Omni-UBench: Benchmarking Omni-Modal LLM-based Guardrails Across Safety, Utility, and Reliability
AI 导读
Omni-UBench 是首个从安全、实用性与可靠性三个维度评估全模态 LLM 护栏的综合基准，涵盖 183k+ 样本、35+ 异构来源，覆盖文本、图像、音频、视频及跨模态组合。对 15 个代表性 OLLM 与全模态护栏的测试显示，现代 OLLM 即使未经护栏专项微调也具备较强内容审核潜力，但现有护栏仍难以识别隐藏有害意图或复杂越狱的跨模态查询，并存在明显的安全幻觉与对文本模式的过度敏感。
正文 · 原文
Abstract:Advances in Omni-modal Large Language Models (OLLMs) have enabled increasingly seamless human-AI interaction, but their unified processing of heterogeneous modalities also expands the attack surface, allowing adversaries to exploit richer input compositions to circumvent safety alignment. To counter such risks, omni-modal guardrails have been proposed to detect and filter unsafe queries across arbitrary modalities. However, their capabilities and limitations remain insufficiently understood due to the absence of a comprehensive evaluation benchmark. In this paper, we present Omni-UBench, the first comprehensive framework for omni-modal safety assessment, evaluating guardrail models along three critical dimensions, i.e., safety, utility, and reliability. Omni-UBench curates over 183k samples drawn from 35+ heterogeneous sources spanning text, image, audio, video, and their cross-modal combinations, annotated with safety labels under a unified risk taxonomy. Benchmarking under Omni-UBench on 15 representative OLLMs and omni-modal guardrails yields several key findings, including: (1) modern OLLMs exhibit strong content moderation potential even without guardrail-specific tuning; (2) existing guardrails still struggle to identify cross-modal queries that conceal harmful intent or involve sophisticated jailbreaks; (3) they further exhibit substantial safety hallucination and over-sensitivity to textual patterns, revealing a deficient understanding of cross-modal safety semantics. Our findings provide valuable insights for developing more trustworthy omni-modal guardrails in future research.
| Subjects: | Cryptography and Security (cs.CR); Computation and Language (cs.CL); Human-Computer Interaction (cs.HC) |
| Cite as: | arXiv:2610.31713 [cs.CR] |
| (or arXiv:2610.31713v1 [cs.CR] for this version) | |
| https://doi.org/10.48550/arXiv.2610.31713 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Junhao Wang [view email]
[v1]
Thu, 22 Oct 2026 04:20:51 UTC (1,602 KB)
来源：arXiv cs.CL · arxiv.org