LapidaryEngine: Fully Conversational Crystal Generation

2026-06-12

AI Summary

Existing text-to-crystal generation methods support only structured inputs and lack bidirectional interaction between natural language and crystal structures, limiting their ability to accommodate the ambiguous and iterative nature of real-world materials design. This work proposes the first conversational crystal generation system, which introduces an intermediate pivot representation to enable, for the first time, bidirectional mapping between text and crystal structures without requiring directly paired training data. Built upon a large language model and integrated with the pivot mechanism, the system supports natural-language-driven interactive design through instruction understanding, crystal generation, and feedback-guided structural editing. Experiments demonstrate significant improvements in alignment between generated crystals and user intent across diverse tasks, including insulator discovery, stability optimization, compositional tuning, and structural editing.

Abstract

The emergence of Large Language Models (LLMs) has inspired the vision of generating bespoke crystal materials directly from natural-language instructions, enabling users to design materials through intuitive, conversational interaction. Existing text-to-crystal generative models represent important early steps toward this goal, but they suffer from two critical limitations: (i) restricted input formats that require highly structured descriptions (e.g., chemical formulas), and (ii) one-directional generation, where models can map text to crystal but cannot perform the inverse. These limitations prevent fully conversational workflows and hinder alignment with users' inherently ambiguous and evolving desiderata. We address these challenges with LapidaryEngine, the first model to support fully conversational crystal generation. LapidaryEngine accepts free-form natural-language requests and performs iterative refinement and editing in a dialogue-like manner. The key innovation is a pivot representation, a third, intermediate form that enables bidirectional translation between text and crystal structures despite the absence of direct paired datasets. Leveraging this pivot allows robust interpretation of user feedback and precise structural control. We demonstrate LapidaryEngine across diverse tasks, including insulator discovery, stability optimization, compositional modification, and structural editing, showcasing its ability to align generated materials with user intent in an interactive manner.

Problem

• text-to-crystal generation
• conversational interaction
• bidirectional translation
• natural-language instructions
• crystal material design

Innovation

• conversational crystal generation
• pivot representation
• bidirectional translation
• natural-language materials design
• iterative refinement