Polymer-Agent: Large Language Model Agent for Polymer Design

2026-01-23

AI Summary

Traditional polymer development relies heavily on inefficient trial-and-error approaches, and existing machine learning tools are often inaccessible to experimental researchers. This work proposes the first closed-loop system that integrates the reasoning capabilities of large language models (LLMs) into polymer design, enabling property prediction, property-driven molecular structure generation, and interactive structural refinement. The framework incorporates constraints based on SMILES sequence generation, synthetic accessibility (SAS), and synthetic complexity (SC Score) to ensure practical feasibility. By embedding these chemically informed priors, the system substantially enhances the synthesizability of generated polymer candidates while offering an intuitive, user-friendly interface for computational assistance. This approach provides experimentalists with actionable design insights and accelerates early-stage polymer discovery.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Cognitive Modeling & Cognitive Systems: Computational Creativity
• Natural Language Processing: (Large) Language Models

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

On-demand Polymer discovery is essential for various industries, ranging from biomedical to reinforcement materials. Experiments with polymers have a long trial-and-error process, leading to use of extensive resources. For these processes, machine learning has accelerated scientific discovery at the property prediction and latent space search fronts. However, laboratory researchers cannot readily access codes and these models to extract individual structures and properties due to infrastructure limitations. We present a closed-loop polymer structure-property predictor integrated in a terminal for early-stage polymer discovery. The framework is powered by LLM reasoning to provide users with property prediction, property-guided polymer structure generation, and structure modification capabilities. The SMILES sequences are guided by the synthetic accessibility score and the synthetic complexity score (SC Score) to ensure that polymer generation is as close as possible to synthetically accessible monomer-level structures. This framework addresses the challenge of generating novel polymer structures for laboratory researchers, thereby providing computational insights into polymer research.

Problem

• Polymer Design
• On-demand Discovery
• Synthetic Accessibility
• Structure-Property Prediction
• Novel Polymer Generation

Innovation

• Large Language Model Agent
• Polymer Design
• Structure-Property Prediction
• Synthetic Accessibility
• SMILES Generation