让智能体通过任务经验持续进化，突破静态推理瓶颈
MetaEvo: A Meta-Optimization Framework for Experience-Driven Agent Evolution
- 分两阶段优化：先提升原则抽象能力，再模块化复用经验
- 在多个推理基准上持续超越基线，迭代中表现稳定提升
- 适合需要长期自我改进的智能体系统开发者
大型语言模型（LLMs）具备强大推理能力，但多数基于LLM的智能体采用静态部署，无法通过任务交互实现进化。现有经验驱动方法常依赖记忆或启发式规则，未真正提升模型学习能力，将其视为被动执行者，导致性能早期停滞且长期改进有限。为此，我们提出MetaEvo——一种两阶段持续智能体进化框架，聚焦于优化模型从任务经验中学习的能力，而非仅关注存储内容。该框架首先通过基于偏好优化提升模型的原则抽象能力，随后在模块化智能体架构中实现这些原则的积累与复用。在多种推理基准上的实验结果表明，MetaEvo consistently优于强基线，并在多轮迭代中保持可靠提升，验证了元优化在使智能体从经验中学习并持续增强推理能力方面的有效性。
原文摘要 · Abstract (English)
Large language models (LLMs) exhibit strong reasoning capabilities, yet most LLM-based agents are statically deployed and unable to improve through task interactions. Existing experience-driven methods often rely on memory or heuristics without enhancing the model's ability to learn, treating it as a passive executor and leading to early performance plateaus and limited long-term improvement. To address this issue, we propose MetaEvo, a two-stage framework for continual agent evolution that focuses on improving how the model learns from tasks experience, rather than solely on what it stores. MetaEvo first applies preference-based optimization to enhance the model's ability of principle abstraction, then enables the accumulation and reuse of these principles within a modular agent architecture. Experimental results on diverse reasoning benchmarks demonstrate that MetaEvo consistently outperforms strong baselines, maintains reliable improvement across iterations. These findings validate the effectiveness of meta-optimization in enabling agents to learn from experience and continually enhance their reasoning capabilities.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。