Saved in:
Bibliographic Details
Main Authors: Kim, Heegyu, Jeon, Taeyang, Choi, Seungtaek, Hong, Ji Hoon, Jeon, Dong Won, Baek, Ga-Yeon, Kwak, Gyeong-Won, Lee, Dong-Hee, Bae, Jisu, Lee, Chihoon, Kim, Yunseo, Choi, Seon-Jin, Park, Jin-Seong, Cho, Sung Beom, Cho, Hyunsouk
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2502.16457
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916656515645440
author Kim, Heegyu
Jeon, Taeyang
Choi, Seungtaek
Hong, Ji Hoon
Jeon, Dong Won
Baek, Ga-Yeon
Kwak, Gyeong-Won
Lee, Dong-Hee
Bae, Jisu
Lee, Chihoon
Kim, Yunseo
Choi, Seon-Jin
Park, Jin-Seong
Cho, Sung Beom
Cho, Hyunsouk
author_facet Kim, Heegyu
Jeon, Taeyang
Choi, Seungtaek
Hong, Ji Hoon
Jeon, Dong Won
Baek, Ga-Yeon
Kwak, Gyeong-Won
Lee, Dong-Hee
Bae, Jisu
Lee, Chihoon
Kim, Yunseo
Choi, Seon-Jin
Park, Jin-Seong
Cho, Sung Beom
Cho, Hyunsouk
contents Materials synthesis is vital for innovations such as energy storage, catalysis, electronics, and biomedical devices. Yet, the process relies heavily on empirical, trial-and-error methods guided by expert intuition. Our work aims to support the materials science community by providing a practical, data-driven resource. We have curated a comprehensive dataset of 17K expert-verified synthesis recipes from open-access literature, which forms the basis of our newly developed benchmark, AlchemyBench. AlchemyBench offers an end-to-end framework that supports research in large language models applied to synthesis prediction. It encompasses key tasks, including raw materials and equipment prediction, synthesis procedure generation, and characterization outcome forecasting. We propose an LLM-as-a-Judge framework that leverages large language models for automated evaluation, demonstrating strong statistical agreement with expert assessments. Overall, our contributions offer a supportive foundation for exploring the capabilities of LLMs in predicting and guiding materials synthesis, ultimately paving the way for more efficient experimental design and accelerated innovation in materials science.
format Preprint
id arxiv_https___arxiv_org_abs_2502_16457
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Towards Fully-Automated Materials Discovery via Large-Scale Synthesis Dataset and Expert-Level LLM-as-a-Judge
Kim, Heegyu
Jeon, Taeyang
Choi, Seungtaek
Hong, Ji Hoon
Jeon, Dong Won
Baek, Ga-Yeon
Kwak, Gyeong-Won
Lee, Dong-Hee
Bae, Jisu
Lee, Chihoon
Kim, Yunseo
Choi, Seon-Jin
Park, Jin-Seong
Cho, Sung Beom
Cho, Hyunsouk
Computation and Language
Materials synthesis is vital for innovations such as energy storage, catalysis, electronics, and biomedical devices. Yet, the process relies heavily on empirical, trial-and-error methods guided by expert intuition. Our work aims to support the materials science community by providing a practical, data-driven resource. We have curated a comprehensive dataset of 17K expert-verified synthesis recipes from open-access literature, which forms the basis of our newly developed benchmark, AlchemyBench. AlchemyBench offers an end-to-end framework that supports research in large language models applied to synthesis prediction. It encompasses key tasks, including raw materials and equipment prediction, synthesis procedure generation, and characterization outcome forecasting. We propose an LLM-as-a-Judge framework that leverages large language models for automated evaluation, demonstrating strong statistical agreement with expert assessments. Overall, our contributions offer a supportive foundation for exploring the capabilities of LLMs in predicting and guiding materials synthesis, ultimately paving the way for more efficient experimental design and accelerated innovation in materials science.
title Towards Fully-Automated Materials Discovery via Large-Scale Synthesis Dataset and Expert-Level LLM-as-a-Judge
topic Computation and Language
url https://arxiv.org/abs/2502.16457