MemEvolve: Meta-Evolution of Agent Memory Systems

Self-evolving memory systems are rapidly reshaping the evolutionary paradigm of large language model (LLM)-based agents. Prior work has predominantly relied on manually engineered memory architectures to store trajectories, distill experience, and synthesize reusable tools, enabling agents to evolve on the fly within environment interactions. However, this paradigm is fundamentally constrained by the \textit{staticity} of the memory system itself: while memory facilitates agent-level evolving, the underlying memory architecture cannot be meta-adapted to diverse task contexts. To address this gap, we propose MemEvolve, a meta-evolutionary framework that jointly evolves agents’ experiential knowledge and their memory architecture, allowing agent systems not only to accumulate experience but also to progressively refine how they learn from it. To ground MemEvolve in prior work and promote openness in future self-evolving systems, we introduce EvolveLab, a unified memory codebase that distills twelve representative memory systems into a modular design space (\textit{encode}, \textit{store}, \textit{retrieve}, \textit{manage}), providing a standardized implementation substrate and a fair experimental arena. Extensive evaluations on four challenging agentic benchmarks show that MemEvolve delivers (i) substantial performance gains, improving frameworks such as SmolAgent and Flash-Searcher by up to $17.06\%$, and (ii) strong cross-task and cross-LLM generalization, yielding memory architectures that transfer effectively across diverse benchmarks and backbones.

AI agents can improve by storing and reusing experience from previous tasks, but most existing systems rely on a fixed memory design chosen by humans. This paper introduces MemEvolve, a framework that lets an agent improve not only the contents of its memory, but also the way its memory is organized and used. MemEvolve evaluates alternative memory designs during task solving and gradually selects more effective ones. We also introduce EvolveLab, a unified codebase for implementing and comparing existing agent memory systems. Across several challenging agent benchmarks, MemEvolve improves performance across different tasks, agent frameworks, and language models, suggesting that adaptive memory design is an important step toward more capable self-improving AI agents.