Predicting Space Groups of Double Perovskites by LLM with Dynamic Few-Shot Learning

2026-08-11

AI Summary

This work addresses the challenge of poor performance on minority space-group classes in double perovskite materials due to severe data imbalance. To this end, the authors propose DyRIS, a framework that integrates large language models with crystallographic prior knowledge. DyRIS employs diversity-enhanced retrieval of contextual examples, dynamic few-shot prompting, and rule-guided reasoning based on B/B' site ordering constraints. It further incorporates quantitative structural descriptors and a dominant space-group bias control mechanism. Evaluated on 3,528 thermodynamically stable samples, DyRIS significantly improves Top-1 accuracy for minority classes—outperforming CrabNet by 3.26%—and achieves state-of-the-art results across all minority-class metrics, with Top-3 accuracy surpassing the strongest baseline.

Abstract

Double perovskites (DPs) offer broad compositional tunability, but predicting the space groups (SGs) of stable structures remains difficult because available datasets are often strongly imbalanced toward dominant SG classes. We refer to dominant SG classes as major SGs and underrepresented classes as minor SGs. We introduce Dynamic and Diversity-enhanced Few-shot Retrieval and Rule-Guided Inference for Space-Group Prediction (DyRIS), an LLM-agent-based framework that predicts ranked SG candidates from a given DP composition. DyRIS uses diversity-enhanced dynamic few-shot prompting to retrieve relevant in-context examples while limiting the dominance of frequently represented SGs. It further incorporates rule-guided inference based on B/B' cation ordering, quantitative indicators, and major-SG bias control to refine and rank the final Top-3 SG candidates. We evaluate DyRIS on 3,528 thermodynamically filtered DP entries and compare it with composition-based and descriptor-based baselines. At a training-data ratio of 0.5, DyRIS achieves competitive overall accuracy while obtaining the best Overall Top-1 macro-F1 score and the best performance across all Minor-SG metrics. DyRIS improves Minor-SG Top-1 accuracy by 3.26 percentage points relative to CrabNet and achieves higher Minor-SG Top-3 accuracy than the strongest PyCaret-based baseline. Ablation studies show that diversity-enhanced retrieval, quantitative indicators, major-SG bias control, and B/B' ordering information each contribute to prediction performance. Additional experiments show that the final rule-guided inference step is not easily replaced by conventional classifier- or ranker-based models. These findings demonstrate the potential of combining retrieval-based LLM reasoning with crystallographic domain knowledge for SG prediction in imbalanced materials datasets.

Problem

• space group prediction
• double perovskites
• imbalanced dataset
• minor space groups
• crystallographic symmetry

Innovation

• dynamic few-shot learning
• space group prediction
• double perovskites
• LLM-agent framework
• imbalanced materials data