让视觉语言模型学会自我纠错，提升光刻缺陷检测精度。
Failure-Aware Refinement of Vision-Language Model for Lithography Defect Detection
- 两阶段框架：先检测，再根据错误样本修正预测结果。
- 在真实数据上显著降低误报和漏检，提升缺陷分类准确率。
- 适合需要高可靠性缺陷检测的半导体制造场景。
半导体光刻检测需可靠识别桥接、毛刺、夹断和污染等微小缺陷。本文提出一种两阶段视觉语言框架，结合初始缺陷检测与预测修正。第一阶段使用LoRA微调Qwen3-VL作为视觉-语言适配器，从光刻图像中预测缺陷数量、类别及归一化边界框。然而直接微调仍会产生常见测试时错误，包括误报、漏检和错误分类。为解决此问题，第二阶段利用第一阶段预测失败案例及其修正标签训练修正模块，使模型能复审并修正初始输出。通过学习失败案例，修正过程超越单阶段微调，显著提升缺陷推断性能。
原文摘要 · Abstract (English)
Semiconductor lithography inspection requires reliable detection of small pattern defects such as bridge, burr, pinch, and contamination. In this study, we propose a two-stage vision-language framework that combines initial defect detection with prediction refinement. In the first stage, Qwen3-VL is fine-tuned with LoRA as a vision-language adapter to predict defect counts, defect categories, and normalized bounding boxes from lithography images. However, direct fine-tuning may still produce common test-time errors, including false positives, missed defects, and incorrect defect types. To address this limitation, the second stage trains a refinement module using first-stage prediction failures and their corrected labels, allowing the model to review and revise initial outputs. By learning from cases where the initial adapter fails, the refinement process improves defect inference beyond single-stage fine-tuning.
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。