收录解读
问题与背景：自动 RTL 优化常在小规模人工退化设计上评测，反馈粗糙且缺少工业 EDA workflow，离真实 PPA 优化较远。
方法与新意：Dr. RTL 在工业 EDA 流程下做 critical-path analysis、并行 RTL rewriting 和 tool-based evaluation，并用 group-relative skill learning 把并行 rewrite 经验蒸馏为可复用 pattern-strategy skill library。
收录意义：这篇对 AI hardware design workflow 很有价值，因为它把 agent 自改 RTL、EDA 工具反馈和技能库积累连成闭环，面向 timing/PPA 的真实工程问题。
局限：工业工具链与 benchmark 细节会影响复现，skill library 的跨项目泛化和安全边界仍需检验。
原始摘要与中文对照
中文对照翻译
大型语言模型（LLMs）的最新进展引发了人们对自动RTL优化以实现更好性能、功耗和面积（PPA）的日益增长的兴趣。然而，现有方法距离真实的RTL优化仍有很大差距：它们的评估往往不切实际，基于手动降级的小规模RTL设计和弱开源工具链；它们的方法也有限，依赖于粗略的设计级反馈和简单的预定义重写规则。为了解决这些局限性，我们提出了Dr. RTL，一个在真实评估环境中进行RTL时序优化的智能体框架，通过可重用的优化技能实现持续自我改进。我们建立了一个真实的评估设置，包含更具挑战性的RTL设计和工业级EDA工作流程。在此设置中，Dr. RTL通过一个多智能体框架进行闭环优化，该框架用于关键路径分析、并行RTL重写和基于工具的评估。我们进一步引入了组相对技能学习，它比较并行RTL重写并将优化经验提炼成一个可解释的技能库。目前，该库包含47个模式-策略条目，用于跨设计重用以改进PPA并加速收敛，并且它可以随着时间的推移持续演进。在20个真实世界RTL设计上进行评估，Dr. RTL在行业领先的商业综合工具基础上，平均实现了21%/17%的WNS/TNS改进，并减少了6%的面积。
原始摘要
Recent advances in large language models (LLMs) have sparked growing interest in automatic RTL optimization for better performance, power, and area (PPA). However, existing methods are still far from realistic RTL optimization: their evaluation is often unrealistic, based on manually degraded, small-scale RTL designs and weak open-source toolchains; their methods are also limited, relying on coarse design-level feedback and simple pre-defined rewriting rules. To address these limitations, we present Dr. RTL, an agentic framework for RTL timing optimization in a realistic evaluation environment, with continual self-improvement through reusable optimization skills. We establish a realistic evaluation setting with more challenging RTL designs and an industrial EDA workflow. Within this setting, Dr. RTL performs closed-loop optimization through a multi-agent framework for critical-path analysis, parallel RTL rewriting, and tool-based evaluation. We further introduce group-relative skill learning, which compares parallel RTL rewrites and distills the optimization experience into an interpretable skill library. Currently, this library contains 47 pattern–strategy entries for cross-design reuse to improve PPA and accelerate convergence, and it can continue evolving over time. Evaluated on 20 real-world RTL designs, Dr. RTL achieves average WNS/TNS improvements of 21%/17% with a 6% area reduction over the industry-leading commercial synthesis tool.