演化環境中的即時推理智能體

摘要

現實世界中的智能體不僅需要做出合乎邏輯的判斷，更需具備時效性決策能力。這要求智能體持續感知動態環境：危險可能突然出現、機會轉瞬即逝、其他智能體不斷行動，而智能體自身的推理過程仍在進行。儘管語言模型推理技術已取得長足進步，現有方法仍未能充分應對這種動態特性。我們提出「即時推理」作為演化環境中智能體的新問題框架，並構建Real-Time Reasoning Gym進行實證研究。我們探討兩種語言模型部署範式：(1) 反應式智能體——採用有限計算資源的語言模型實現快速響應；(2) 規劃式智能體——允許擴展推理計算以處理複雜問題。實驗表明，即使最先進的模型在兩種範式下都難以兼顧邏輯嚴謹性與時效性。為突破此限制，我們提出AgileThinker框架，能同步運用兩種推理範式。隨著任務難度與時間壓力提升，AgileThinker始終優於單一推理範式的智能體，有效平衡推理深度與響應延遲。本研究確立即時推理作為開發實用智能體的關鍵測試平台，為時序約束型AI系統研究奠定基礎，指明了實現具備實時能力智能體的發展路徑。

English

Agents in the real world must make not only logical but also timely judgments. This requires continuous awareness of the dynamic environment: hazards emerge, opportunities arise, and other agents act, while the agent's reasoning is still unfolding. Despite advances in language model reasoning, existing approaches fail to account for this dynamic nature. We introduce real-time reasoning as a new problem formulation for agents in evolving environments and build Real-Time Reasoning Gym to demonstrate it. We study two paradigms for deploying language models in agents: (1) reactive agents, which employ language models with bounded reasoning computation for rapid responses, and (2) planning agents, which allow extended reasoning computation for complex problems. Our experiments show that even state-of-the-art models struggle with making logical and timely judgments in either paradigm. To address this limitation, we propose AgileThinker, which simultaneously engages both reasoning paradigms. AgileThinker consistently outperforms agents engaging only one reasoning paradigm as the task difficulty and time pressure rise, effectively balancing reasoning depth and response latency. Our work establishes real-time reasoning as a critical testbed for developing practical agents and provides a foundation for research in temporally constrained AI systems, highlighting a path toward real-time capable agents.

演化環境中的即時推理智能體

Real-Time Reasoning Agents in Evolving Environments

摘要

Support