UniPCB: A Unified Vision-Language Benchmark for Open-Ended PCB Quality Inspection

TL;DR

This paper introduces UniPCB, a comprehensive benchmark for PCB quality inspection using vision-language models, and presents PCB-GPT, a specialized model trained on a new dataset that significantly outperforms existing models in defect localization.

Contribution

The paper creates the first unified PCB inspection benchmark and develops PCB-GPT, a new MLLM trained with a curriculum learning approach tailored for PCB inspection tasks.

Findings

PCB-GPT more than doubles defect localization performance.

Existing MLLMs perform poorly on domain-specific PCB tasks.

The benchmark standardizes evaluation across diverse PCB inspection scenarios.

Abstract

Multimodal Large Language Models (MLLMs) show promise for general industrial quality inspection, but fall short in complex scenarios, such as Printed Circuit Board (PCB) inspection. PCB inspection poses unique challenges due to densely packed components, complex wiring structures, and subtle defect patterns that require specialized domain expertise. However, a high-quality, unified vision-language benchmark for quantitatively evaluating MLLMs across PCB inspection tasks remains absent, stemming not only from limited data availability but also from fragmented datasets and inconsistent standardization. To fill this gap, we propose UniPCB, the first unified vision-language benchmark for open-ended PCB quality inspection. UniPCB is built via a systematic pipeline that curates and standardizes data from disparate sources across three annotated scenarios. Furthermore, we introduce PCB-GPT, an MLLM trained on a new instruction dataset generated by this pipeline, utilizing a novel progressive curriculum that mimics the learning process of human experts. Evaluations on the UniPCB benchmark show that while existing MLLMs falter on domain-specific tasks, PCB-GPT establishes a new baseline. Notably, it more than doubles the performance on fine-grained defect localization compared to the strongest competitors, with significant advantages in localization and analysis. We will release the instruction data, benchmark, and model to facilitate future research.