General Agentic Memory Via Deep Research

TL;DR

This paper introduces General Agentic Memory (GAM), a dynamic memory framework for AI agents that improves task performance by creating optimized, context-specific memory at runtime, addressing limitations of static memory systems.

Contribution

GAM is a novel framework combining lightweight historical memory with a universal store, enabling scalable, context-aware memory retrieval guided by reinforcement learning.

Findings

GAM significantly outperforms existing memory systems in task completion.

GAM effectively leverages large language models for agentic memory tasks.

End-to-end optimization enhances GAM's performance across scenarios.

Abstract

Memory is critical for AI agents, yet the widely-adopted static memory, aiming to create readily available memory in advance, is inevitably subject to severe information loss. To address this limitation, we propose a novel framework called \textbf{general agentic memory (GAM)}. GAM follows the principle of "\textbf{just-in time (JIT) compilation}" where it focuses on creating optimized contexts for its client at runtime while keeping only simple but useful memory during the offline stage. To this end, GAM employs a duo-design with the following components. 1) \textbf{Memorizer}, which highlights key historical information using a lightweight memory, while maintaining complete historical information within a universal page-store. 2) \textbf{Researcher}, which retrieves and integrates useful information from the page-store for its online request guided by the pre-constructed memory. This design allows GAM to effectively leverage the agentic capabilities and test-time scalability of frontier large language models (LLMs), while also facilitating end-to-end performance optimization through reinforcement learning. In our experimental study, we demonstrate that GAM achieves substantial improvement on various memory-grounded task completion scenarios against existing memory systems.