Robuta

https://www.utwente.nl/en/eemcs/scs/education/assignments/finished-assignments/master/20260129-cyberbench-a-taxonomy-dataset-and-framework-for-automated-benchmarking-of-direct-security-risks-in-llm-based-systems/ CyberBench: A Taxonomy, Dataset and Framework for Automated Benchmarking of Direct Security Risks... automated benchmarkingsecurity riskscyberbenchtaxonomydataset