Preprint
Aug 2026
$A^2E$ : An End-to-End Agent Auditing Engine
Experiments conducted with A^2E (Agent Auditing Engine) reveal that model-harness combinations exhibit substantial performance variation across different types of tasks, and that no single combination consistently outperforms all others across every task.
Haoning Wang, Mingxun Zhang, Chenyue Yu et al.
· 0 citations