作者位次 #5/8ACL 2026Findings· Findings of ACL
A gentic E val: Toward Agentic and Self-Evolving Safety Evaluation of Large Language Models
Yixu Wang · Xin Wang · Yang Yao · Xinyuan Li · Xibang Yang · Yan Teng · Xingjun Ma · Yingchun Wang
摘要
The rapid integration of Large Language Models (LLMs) into high-stakes domains necessitates reliable safety and compliance evaluation. However, existing static benchmarks are ill-equipped to address the dynamic nature of AI risks and evolving regulations, creating a critical safety gap. This pape...