evaluation-resource · Source-linked discovery
MM-SafetyBench
Evaluates multimodal model safety under image-text attack settings.
OriginMM-SafetyBench paper authorsTopicsSafeguards · MultimodalStatuscatalogued
Can support
Not independently assessed by FronteraEval yet.
Cannot support by itself
No inference beyond the upstream source should be made until the protocol is reviewed.