The rise of Large Reasoning Models (LRMs) signifies a paradigm shift toward advanced computational reasoning. Yet, this progress disrupts traditional agent frameworks, traditionally anchored by execution-oriented Large Language Models (LLMs). To explore this transformation, we propose the LaRMA framework, encompassing nine tasks across Tool Usage, Plan Design, and Problem Solving, assessed with three top LLMs (e.g., Claude3.5-sonnet) and five leading LRMs (e.g., DeepSeek-R1). Our findings address four research questions: LRMs surpass LLMs in reasoning-intensive tasks like Plan Design, leveraging iterative reflection for superior outcomes; LLMs excel in execution-driven tasks such as Tool Usage, prioritizing efficiency; hybrid LLM-LRM configurations, pairing LLMs as actors with LRMs as reflectors, optimize agent performance by blending execution speed with reasoning depth; and LRMs' enhanced reasoning incurs higher computational costs, prolonged processing, and behavioral challenges, including overthinking and fact-ignoring tendencies. This study fosters deeper inquiry into LRMs' balance of deep thinking and overthinking, laying a critical foundation for future agent design advancements.

翻译：大规模推理模型（LRMs）的兴起标志着计算推理能力向高级范式转变。然而，这一进展对传统以执行为导向的大语言模型（LLMs）为基石的智能体框架构成了冲击。为探究这一变革，我们提出LaRMA评估框架，涵盖工具使用、规划设计和问题解决三大领域的九项任务，并评估了三款顶尖LLMs（如Claude3.5-sonnet）与五款领先LRMs（如DeepSeek-R1）的表现。研究结果回应了四个核心问题：在规划设计等推理密集型任务中，LRMs凭借迭代反思机制显著优于LLMs；LLMs在工具使用等执行驱动型任务中表现更佳，注重效率优先；采用LLM作为执行器与LRM作为反思器的混合架构，能融合执行速度与推理深度，从而优化智能体性能；LRMs增强的推理能力伴随更高计算成本、更长处理时间及行为挑战，包括过度思考与忽视事实的倾向。本研究推动了对LRMs深度思考与过度思考平衡机制的深入探讨，为未来智能体设计的进步奠定了关键基础。