SGLayout:用 Schema 引导闭环修正提升 LLM 参数化版图生成
SGLayout: Schema-Guided Closed-Loop Refinement for LLM-Driven Parameterized Layout Generation
做芯片版图自动化的可以看看这篇。它用 schema 把 DRC 检查出的几何违例直接映射回代码位置来修,112 个 GDSFactory 任务上通过率从 50.89% 拉到 76.79%,比一版版盲改的迭代方式靠谱多了。
论文提出 SGLayout 框架,针对 LLM 在参数化版图生成中缺乏空间感知、一次性生成不可靠的问题。框架在代码执行后通过结构化 schema 建立代码实例与几何属性的双向映射,打通几何评估结果到代码修复之间的链路。评估沿两个维度展开:意图合规采用定量结构检查加视觉-语言定性评估,规则合规采用 DRC 设计规则检查,所有违例再经 schema 回投到具体代码位置进行精准修复。
在基于 GDSFactory 构建的 112 个参数化版图生成任务基准上,SGLayout 达到 76.79% 的通过率,最强基线为 50.89%,提升约 26 个百分点。该闭环迭代持续到版图收敛为止,模拟工程师 review 的流程。对 EDA 与芯片设计自动化方向的开发者而言,schema 作为代码空间与几何空间之间的桥梁这一设计思路,也可迁移到其他需要几何反馈的代码生成任务。
SGLayout: Schema-Guided Closed-Loop Refinement for LLM-Driven Parameterized Layout Generation
Large Language Models (LLMs) have shown strong potential in parameterized layout generation via tool-augmented code synthesis. However, LLMs lack spatial awareness, making one-shot generation unreliable. While layout quality can be evaluated with established techniques, a critical gap remains: geometric-domain evaluation outcomes cannot be directly translated into actionable code-level fixes, trapping iterative refinement in blind refactoring. We propose SGLayout, a self-refining framework that closes this gap via a schema-guided mechanism. Upon code execution, a structured schema captures the correspondence between each code-level instance and its geometric attributes, establishing a bidirectional link between code space and geometry space. Emulating engineer review practices, SGLayout evaluates layouts along two complementary dimensions: intent compliance through quantitative structural checking and qualitative vision-language assessment, and rule compliance through DRC. All identified violations are then back-projected through the schema onto precise code targets, enabling surgical repair. This closed-loop process iterates until layout convergence. On a curated benchmark of 112 parameterized layout generation tasks using GDSFactory, SGLayout achieves a 76.79% pass rate, compared with 50.89% for the strongest baseline.
This report is categorized under AITOP's "Papers" desk. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include SGLayout, LLM, GDSFactory, EDA.
Source attribution: curated from arXiv cs.AI (@Siyang Chen, Yuan Pu, Yuhan Qin, Zhaotan Lin, Siyuan Liang, Zhuolun He, Lining Zhang, Bei Yu).