Foundational Model for Electron Micrograph Analysis: Instruction-Tuning Small-Scale Language-and-Vision Assistant for Enterprise Adoption

TL;DR

This paper presents a small-scale, instruction-tuned multimodal model for semiconductor electron micrograph analysis, enabling accurate, private, and cost-effective microscopic image understanding for enterprise use.

Contribution

It introduces a novel instruction-tuning framework with knowledge distillation for semiconductor microscopy, reducing reliance on costly expert annotations and enabling enterprise customization.

Findings

Outperforms traditional microscopic analysis methods.

Adapts effectively to data distribution shifts.

Supports high-throughput screening in semiconductor manufacturing.

Abstract

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.