Story
arxiv_llm_reliability · Oct 22, 2026 · paper
Source brief
Omni-UBench: Benchmarking Omni-Modal LLM-based Guardrails Across Safety, Utility, and Reliability
arxiv.orgOct 22, 2026
original source linked
In brief
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 adversar...
Feed lens
evaluation