Foundational Model for Electron Micrograph Analysis: Instruction-Tuning Small-Scale Language-and-Vision Assistant for Enterprise Adoption
This addresses the understudied domain of semiconductor imaging for manufacturing optimization, though it appears incremental as it adapts existing multimodal methods to a specific application.
The paper tackles the problem of semiconductor electron microscopy image analysis by introducing a small-scale multimodal framework (MAEMI) that uses vision-language instruction tuning and knowledge distillation, eliminating the need for expensive human-annotated datasets and enabling enterprise adoption on low-cost hardware.
Semiconductor imaging and analysis are critical yet understudied in deep learning, limiting our ability for precise control and optimization in semiconductor manufacturing. We introduce a small-scale multimodal framework for analyzing semiconductor electron microscopy images (MAEMI) through vision-language instruction tuning. We generate a customized instruction-following dataset using large multimodal models on microscopic image analysis. We perform knowledge transfer from larger to smaller models through knowledge distillation, resulting in improved accuracy of smaller models on visual question answering (VQA) tasks. This approach eliminates the need for expensive, human expert-annotated datasets for microscopic image analysis tasks. Enterprises can further finetune MAEMI on their intellectual data, enhancing privacy and performance on low-cost consumer hardware. Our experiments show that MAEMI outperforms traditional methods, adapts to data distribution shifts, and supports high-throughput screening.