Investigate-Consolidate-Exploit: A General Strategy for Inter-Task Agent Self-Evolution
Cheng Qian Shihao Liang Yujia Qin Yining Ye Xin Cong Yankai Lin Yesai Wu Zhiyuan Liu Maosong Sun
Abstract
This paper introduces Investigate-Consolidate-Exploit (ICE), a novel strategy for enhancing the adaptability and flexibility of AI agents through inter-task self-evolution. Unlike existing methods focused on intra-task learning, ICE promotes the transfer of knowledge between tasks for genuine self-evolution, similar to human experience learning. The strategy dynamically investigates planning and execution trajectories, consolidates them into simplified workflows and pipelines, and exploits them for improved task execution. Our experiments on the XAgent framework demonstrate ICE’s effectiveness, reducing API calls by as much as 80% and significantly decreasing the demand for the model’s capability. Specifically, when combined with GPT-3.5, ICE’s performance matches that of raw GPT-4 across various agent tasks. We argue that this self-evolution approach represents a paradigm shift in agent design, contributing to a more robust AI community and ecosystem, and moving a step closer to full autonomy.
中文速览
AI智能体（AI agent）在跨任务场景中无法像人类一样把过去的经验迁移到新任务上，每次都要从头摸索，既费时又费算力。ICE（Investigate-Consolidate-Exploit）策略将智能体的任务规划记录和执行轨迹分别提取、压缩，再存入记忆库：先"调查"哪些经验值得保留，再"整合"成标准化的线性工作流和有限自动机流水线，最后在处理新任务时"利用"这些经验来直接指导规划或自动执行重复步骤。在XAgent框架上的实验表明，ICE最多可将API调用次数降低80%，并大幅降低对基础模型能力的依赖——配合GPT-3.5时，其表现可媲美原始GPT-4。这一跨任务自进化范式让智能体像人一样从经验中成长，为构建更高效、更自主的AI系统提供了切实可行的路径。
原文 arXiv:2401.13996;中英对照 + 大白话阅读 https://aha.fim.ai/paper/2401.13996v1