GlazyBench: A Benchmark for Ceramic Glaze Property Prediction and Image Generation

2026-05-07

AI Summary

This work addresses the longstanding reliance on costly trial-and-error in ceramic glaze development by introducing GlazyBench, the first large-scale benchmark dataset for AI-driven glaze design. Comprising 23,148 real-world glaze formulations, GlazyBench enables predictive modeling of fired surface properties from raw materials and supports attribute-conditioned generation of photorealistic glaze images. The study establishes the first standardized evaluation framework for this domain, systematically evaluating traditional machine learning, large language models, and multimodal generative approaches on both property prediction and image synthesis tasks. Experimental results demonstrate the potential of data-driven methods while highlighting current limitations in modeling complex ceramic firing processes, thereby paving a new pathway for AI-assisted design of traditional materials.

Abstract

Developing ceramic glazes is a costly, time-consuming process of trial and error due to complex chemistry, placing a significant burden on independent artists. While recent advances in multimodal AI offer a modern solution, the field lacks the large-scale datasets required to train these models. We propose GlazyBench, the first dataset for AI-assisted glaze design. Comprising 23,148 real glaze formulations, GlazyBench supports two primary tasks: predicting post-firing surface properties, such as color and transparency, from raw materials, and generating accurate visual representations of the glaze based on these properties. We establish comprehensive baselines for property prediction using traditional machine learning and large language models, alongside image generation benchmarks using deep generative and large multimodal models. Our experiments demonstrate promising yet challenging results. GlazyBench pioneers a new research direction in AI-assisted material design, providing a standardized benchmark for systematic evaluation.

Problem

• ceramic glaze
• property prediction
• image generation
• multimodal AI
• material design

Innovation

• GlazyBench
• ceramic glaze prediction
• multimodal AI
• material design
• image generation