ailor/AdvBench-omni download history
ailor/AdvBench-omni is an any to any dataset on the Hugging Face Hub. In the last 30 days it was downloaded 347 times (41 in the last 7 days), and 10,974 times in total. It ranks #43,845 among datasets by monthly downloads.
AdvBench-Omni Paper | Code AdvBench-Omni is a dataset constructed to evaluate and reveal safety vulnerabilities in Omni-modal Large Language Models (OLLMs). It is based on a modality-semantics decoupling principle to study how OLLMs handle cross-modal safety risks and conflicts.