benchmark · first seen 13 Oct 2026
agent privacy evaluation
Benchmarking whether tool-using language agents leak sensitive user data across application environments.
Mentions per week
PapersCodeDiscussion
Earliest sources
- AgentCIScope: Grounded Contextual Integrity Evaluation for LLM AgentsPapers · 13 Oct 2026 · Chengquan Guo, Zhaorun Chen, Zinan Lin, Xiaodong Liu, Bo Li
Latest sources
- AgentCIScope: Grounded Contextual Integrity Evaluation for LLM AgentsPapers · 13 Oct 2026 · Chengquan Guo, Zhaorun Chen, Zinan Lin, Xiaodong Liu, Bo Li