发表机构
Fudan University; Chinese University of Hong Kong（复旦大学; 香港中文大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
StateTune将LLM辅助的EDA调优改为有状态闭环过程，以持久优化记忆为核心，结合EHVI引导的提升策略，在六个基准模块上性能优于多个基线，验证了设计有效性。
AI 中文摘要
EDA流程参数调优对结果质量（QoR）至关重要，但参数空间庞大、耦合紧密，完整评估成本过高。现有大语言模型（LLM）辅助调优器主要将LLM作为带有临时工作上下文的外部提议者，我们提出StateTune，将LLM辅助的EDA调优重新表述为有状态的闭环过程。其优化器状态是一种带类型、经证据门控的持久优化记忆，每次评估后都会更新，并在候选生成和预算分配间共享。在此优化器状态基础上，采用期望超体积改进（EHVI）引导、感知运行时间的提升策略，按每单位运行时间成本的期望帕累托前沿增益对快速阶段候选进行排名。在Cadence工业流程上，针对六个基准模块（两个技术节点×三种设计），与包括LLM+检索增强生成（RAG）和基于偏好的贝叶斯优化（BO）调优器在内的五个基线对比，StateTune在所有六个基准模块上均实现最强的最终超体积，在整个矩阵中展现出前沿质量的稳定提升；在最坏负 slack（WNS）、面积和功耗方面也达到或超过同一组中最强的基线。消融实验显示，持久记忆是最大贡献者：移除它会损失58.5%的超体积。对证据门控敏感性、记忆污染、跨设计迁移和三种子重现性（六个模块中有五个的变异系数CV<7%）的专门分析进一步验证了该记忆设计。
英文摘要
EDA flow parameter tuning is critical for quality-of-results~(QoR), yet the parameter space is large, tightly coupled, and full evaluations are prohibitively expensive. Prior LLM-assisted tuners mainly use the LLM as an external proposer with transient working context; we instead present \textbf{StateTune}, which reformulates LLM-assisted EDA tuning as a closed-loop, state-carrying process. Its optimizer state is a typed, evidence-gated \emph{persistent optimization memory} that is updated by every evaluation and shared between candidate generation and budget allocation. On top of this optimizer state, an expected hypervolume improvement (EHVI)-guided, runtime-aware promotion policy ranks quick-stage candidates by expected Pareto frontier gain per unit of runtime cost. Evaluated on a Cadence industrial flow across six benchmark blocks (two technology nodes \(\times\) three designs), against five baselines including LLM+retrieval-augmented generation (RAG) and preference-based Bayesian optimization (BO) tuners, StateTune achieves the strongest final hypervolume on all six benchmark blocks, showing a stable improvement in frontier quality across the full matrix; it also matches or surpasses the strongest baselines on worst negative slack (WNS), area, and power across the same set. Ablation shows persistent memory is the largest contributor: removing it costs 58.5\% of the hypervolume. Dedicated analyses of evidence-gating sensitivity, memory poisoning, cross-design transfer, and three-seed reproducibility (CV\,\(<\)\,7\% on five of six blocks) further validate the memory design.
CommentsAccepted to the 2026 IEEE/ACM International Conference on Computer-Aided Design (ICCAD 2026)
