AI RESEARCH
Reasoning and Tool-use Compete in Agentic RL:From Quantifying Interference to Disentangled Tuning
arXiv CS.AI
•
ArXi:2602.00994v2 Announce Type: replace Agentic Reinforcement Learning (ARL) trains large language models to interleave reasoning with external tool execution to solve complex tasks. Most existing ARL methods train a single set of parameters to both reasoning and tool-use behaviors, implicitly assuming that joint