发表机构
Southeast University; National Center of Technology Innovation for EDA; NVIDIA Corporation; Fudan University; Nanjing University of Posts and Telecommunications; Institute of Computing Technology, Chinese Academy of Sciences; Peking University（东南大学; 国家EDA技术创新中心; 英伟达公司; 复旦大学; 南京邮电大学; 中国科学院计算技术研究所; 北京大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
MicroEvo是结合LLM与MCTS的知识引导框架，用于多目标微架构优化，可提升帕累托前沿质量与搜索效率，具备工业级可扩展性。
AI 中文摘要
微架构设计空间探索面临搜索空间庞大、PPA评估成本高昂的问题，仅能为设计决策提供有限的仿真预算。现有方法采用无感知搜索，未考虑微架构依赖关系且无法从迭代搜索中有效学习，导致评估资源浪费和帕累托收敛效果弱。本文提出MicroEvo，一种知识引导框架，将现成的大语言模型（LLM）与蒙特卡洛树搜索（MCTS）结合用于多目标微架构优化。MicroEvo包含四个关键组件：LLM驱动的进化算子、平衡帕累托贡献与多样性的帕累托感知树策略、提取并复用优化见解的主动知识积累机制，以及在线调整搜索行为的状态感知指令。实验表明，MicroEvo的帕累托前沿质量相比NSGA-II提升最高达36.2%，搜索效率提高10.6倍，且对复杂工业级核心展现出强可扩展性。代码仓库可访问：this https URL
英文摘要
Microarchitecture design space exploration suffers from expansive search spaces and expensive PPA evaluation, leaving only a small simulation budget for design decision-making. Existing methods perform blind search without considering microarchitectural dependencies and fail to learn from the iterative search effectively, leading to wasted evaluations and weak Pareto convergence. In this paper, we propose MicroEvo, a knowledge-guided framework that couples off-the-shelf LLMs with Monte Carlo Tree Search (MCTS) for multi-objective microarchitecture optimization. MicroEvo combines LLM-driven evolutionary operators, a Pareto-aware tree policy that balances Pareto contribution and diversity, an active knowledge accumulation mechanism that extracts and reuses optimization insights, and state-aware directives that adapt the search behavior online. Experiments show that MicroEvo improves Pareto-front quality by up to 36.2% over NSGA-II and achieves 10.6x higher search efficiency, and also demonstrates strong scalability to a complex industrial-scale core. The code repository is available at: https://github.com/GEAR-SEU/MicroEvo-ICCAD-26.
CommentsAccepted by ICCAD 2026
