让AI学会自己摸索电子设计工具的隐藏用法，自动写对脚本。
ZhuLong: Execution-Grounded LLM Agent for EDA Scripting with Offline API Self-Exploration
- 通过离线实验推断未文档化的工具接口行为，结合真实执行验证
- 在真实商业环境中达到78.5%的脚本正确率，远超纯AI基线
- 适合需要自动化芯片设计脚本的工程师和研究者
使用特定工具、常无文档的EDA API进行脚本编写仍是长期存在的瓶颈，现有LLM无法有效解决。本文提出ZhuLong，一个基于执行反馈的LLM编程代理，适用于PyAether和SKILL，融合了API检索、文档查阅与沙箱执行，通过离线API自探索机制，借助反事实实验推断未文档化接口的行为。我们在包含158个真实任务的EDA-Eval-PyAether基准上评估，完整系统在商业Empyrean Aether环境中达到78.5% Pass@1，显著优于纯LLM基线（23.6%）。消融实验表明沙箱执行是主要性能驱动因素（移除后下降41.2个百分点），自探索机制额外提升3.2个百分点准确率，并将每任务工具调用减少22.1%。在20个涉及未保存版图和原理图的交互任务中，PyAether达60.0% Pass@1，SKILL达50.0%。
原文摘要 · Abstract (English)
EDA scripting with tool-specific, often undocumented APIs remains a long-tail bottleneck that existing LLMs fail to address. This paper presents ZhuLong, an execution-grounded LLM coding agent for PyAether and SKILL that combines API retrieval, documentation inspection, and sandbox execution via unified MCP tools, augmented by an offline API self-exploration mechanism that infers undocumented API behaviors through counterfactual experimentation. We evaluate ZhuLong on EDA-Eval-PyAether, a benchmark of 158 real-world tasks with assertion-based execution, where the complete system achieves 78.5% Pass@1 in the commercial Empyrean Aether environment, substantially outperforming a pure LLM baseline (23.6%). Ablation studies identify sandbox execution as the dominant performance driver (41.2 pp drop when removed), with the self-exploration mechanism contributing an additional 3.2 pp accuracy gain and a 22.1% reduction in per-task tool calls. On 20 interactive tasks involving unsaved layouts and schematics, ZhuLong achieves 60.0% Pass@1 for PyAether and 50.0% for SKILL.
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。