AI RESEARCH
On Effectiveness and Efficiency of Agentic Tool-calling and RL Training
arXiv CS.AI
•
ArXi:2606.00135v1 Announce Type: cross Tool-calling is a central component of modern large language model (LLM) agents, equipping them with skills beyond their parametric knowledge. This paper studies tool-calling along two complementary axes: effectiveness, i.e., how this capability is measured, and efficiency, i.e., how it is learned.