1 paper · 1 filter
Hassan Hamad, Yingru Xu, Liang Zhao +2
Tool-augmented large language models (LLMs) are increasingly employed in real-world applications, but tool usage errors still hinder their reliability. We introduce ToolCritic, a d…