跳到正文
返回论文列表
cs.CE提交于 已译

基于少样本 XSEM 数据与混合搜索的 Bosch 刻蚀模拟器实仿真校准

Real-to-Sim Calibration of Bosch Etching Simulators with Few-Shot XSEM Data and Hybrid Search

Xinyu Jia · Jainil Shah · Rahul Tripathi · Vijay Gupta · Zhihong Chen

中文摘要

Bosch 工艺广泛用于 MEMS 和 TSV 制造,但真实测量稀疏且成本高昂,缺乏校准时难以准确预测刻蚀结果。该少样本实仿真(real-to-sim)校准框架将模拟器参数与直接工艺变量解耦,从稀疏测量推断特定配方的模拟器参数向量,并在留出工艺条件下可靠预测刻蚀深度和侧壁角。研究将 Bosch 轮廓预测建模为少样本实仿真校准问题,以少量真实 XSEM 测量识别特定的 ViennaPS 参数向量 z_r^*,使模拟与实测刻蚀深度、侧壁角的差异最小化。搜索过程结合局部探索、LLM 引导候选生成和基于 XGBoost 的代理模型筛选,最终仍由 ViennaPS 仿真验证每个候选。在两种 Bosch 刻蚀配方上的实验表明,该框架能准确预测留出结果,归一化全局误差分别为 0.340 和 0.249。

关键要点

  1. 01真实测量稀疏且昂贵,Bosch 刻蚀结果难以直接准确预测
  2. 02将配方特定的 ViennaPS 参数与直接工艺变量解耦校准
  3. 03结合局部探索、LLM 候选生成与 XGBoost 代理筛选
  4. 04以少量真实 XSEM 数据预测留出刻蚀深度和侧壁角
  5. 05两种配方的归一化全局误差分别为 0.340 和 0.249

解读

尚无解读。

原始英文摘要

arXiv:2610.31904v1 Announce Type: cross Abstract: The Bosch process is widely used in MEMS and TSV fabrication, but accurate prediction of etching outcomes remains difficult without calibration against real measurements, which are sparse and costly to obtain. This work presents a few-shot real-to-sim calibration framework that decouples simulator parameters from direct process variables to infer a recipe-specific simulator parameter vector from sparse measurements, enabling reliable prediction of etch depth and sidewall angle under held-out process conditions. We first formulate Bosch profile prediction as a few-shot real-to-sim calibration problem, where a small set of real XSEM measurements is used to identify a recipe-specific ViennaPS parameter vector $z_r^*$ that minimizes the discrepancy between simulated and measured etch depth and sidewall angle. To search this parameter space efficiently, the framework combines local exploration, LLM-guided candidate generation, and XGBoost-based surrogate screening, while retaining ViennaPS simulation as the final verification step for each selected candidate. Experiments on two Bosch etching recipes show that the proposed framework accurately predicts held-out etching results, achieving normalized global errors of 0.340 and 0.249 for the 2 recipes respectively.

同方向论文 · cs.CE

查看全部 →