Showing cs.CRShow all
3 papers · 1 filter
cs.CR2026
EvoSafeHarness: Evolving Model- and Domain-Specific Harnesses for Securing Agents
Nanxi Li, Yingzi Ma, Yulong Cao +7
Large language model (LLM) agents can translate natural-language requests into real-world actions, creating safety risks from both indirect prompt injections and directly harmful r…
cs.CR2026
LPG: Balancing Efficiency and Policy Reasoning in Latent Policy Guardrails
Nanxi Li, Zhengyue Zhao, Chaowei Xiao
Guardrails are a critical safety layer for modern AI systems, but their operating regime is changing. As LLMs are deployed as customized assistants, safety policies are increasingl…
cs.CR2025
PRISM: Robust VLM Alignment with Principled Reasoning for Integrated Safety in Multimodality
Nanxi Li, Zhengyue Zhao, G. Edward Suh +2
Safeguarding vision-language models (VLMs) is a critical challenge, as existing methods often suffer from over-defense, which harms utility, or rely on shallow alignment, failing t…