MemoBrain：將執行記憶體作為推理驅動的智能大腦

MemoBrain: Executive Memory as an Agentic Brain for Reasoning

January 12, 2026

作者: Hongjin Qian, Zhao Cao, Zheng Liu

cs.AI

摘要

在工具增強型智慧體框架中，複雜推理本質上具有長期性特徵，這會導致推理軌跡和瞬態工具產物不斷累積，從而對大型語言模型的有限工作上下文造成壓力。若缺乏顯性記憶機制，此類累積會破壞邏輯連續性並削弱任務對齊效果。這使記憶不再僅是輔助性的效率優化項，而成為維持長期連貫目標導向推理的核心組件。 我們提出MemoBrain——一種面向工具增強型智慧體的執行記憶模型，它能構建推理步驟間的依賴感知記憶，捕捉關鍵中間狀態及其邏輯關聯。MemoBrain作為推理智慧體的協同駕駛員，在無需阻斷執行的前提下組織推理進程，並主動管理工作上下文。具體而言，該模型會修剪無效步驟、折疊已完成子軌跡，並在固定上下文預算下保留緊湊的高顯著性推理主幹。這些機制共同實現了對推理軌跡的顯性認知控制，而非被動的上下文堆積。 我們在GAIA、WebWalker和BrowseComp-Plus等具有挑戰性的長期基準測試中評估MemoBrain，結果顯示其相較於強基線模型均能實現持續效能提升。

English

Complex reasoning in tool-augmented agent frameworks is inherently long-horizon, causing reasoning traces and transient tool artifacts to accumulate and strain the bounded working context of large language models. Without explicit memory mechanisms, such accumulation disrupts logical continuity and undermines task alignment. This positions memory not as an auxiliary efficiency concern, but as a core component for sustaining coherent, goal-directed reasoning over long horizons. We propose MemoBrain, an executive memory model for tool-augmented agents that constructs a dependency-aware memory over reasoning steps, capturing salient intermediate states and their logical relations. Operating as a co-pilot alongside the reasoning agent, MemoBrain organizes reasoning progress without blocking execution and actively manages the working context. Specifically, it prunes invalid steps, folds completed sub-trajectories, and preserves a compact, high-salience reasoning backbone under a fixed context budget. Together, these mechanisms enable explicit cognitive control over reasoning trajectories rather than passive context accumulation. We evaluate MemoBrain on challenging long-horizon benchmarks, including GAIA, WebWalker, and BrowseComp-Plus, demonstrating consistent improvements over strong baselines.