Showing 2026Show all
2 papers · 1 filter
eess.SY2026
Weak and Strong Data Informativity with Structural Priors for System Analysis
Yuan Zhang, Wenxuan Xu, Yun Feng +2
When measurement data are insufficient for unique system identification, data informativity evaluates whether a dataset determines a property across all systems consistent with the…
cs.CR2026
Understanding In-Context Multimodal Jailbreaks via Posterior Reweighting
Xu Zhang, Dev Mistry, Xiang Xu +1
In-context learning (ICL) jailbreaks reveal a critical vulnerability in multimodal large language models (MLLMs): harmful demonstrations in the prompt can induce unsafe outputs wit…