Autonomous Multi-Agent AI for High-Throughput Polymer Informatics: From Property Prediction to Generative Design Across Synthetic and Bio-Polymers
🤖 AI Summary
This work proposes the first multi-agent AI ecosystem tailored to the full polymer research lifecycle (PRL), addressing the limitations of traditional polymer development that relies heavily on manual expertise and lacks an efficient, unified intelligent framework for high-throughput property prediction and generative design. The system integrates literature comprehension, tool invocation, code generation, and metacognitive self-evaluation to enable end-to-end automation. Innovatively combining multi-agent collaboration with a metacognitive mechanism, it leverages the DeepSeek large language model, graph neural networks (PolyGNN), and multimodal representations to support cross-modal biopolymer modeling, uncertainty quantification, and self-optimizing strategies. Evaluated on a test set of 1,251 polymers, the framework achieves R² values of 0.75–0.91 for key properties such as glass transition temperature (Tg), with a single inference requiring only 16.3 seconds and 0.1 GPU hours (approximately $0.08), substantially outperforming existing methods.
Application Category
📝 Abstract
We present an integrated multiagent AI ecosystem for polymer discovery that unifies high-throughput materials workflows, artificial intelligence, and computational modeling within a single Polymer Research Lifecycle (PRL) pipeline. The system orchestrates specialized agents powered by state-of-the-art large language models (DeepSeek-V2 and DeepSeek-Coder) to retrieve and reason over scientific resources, invoke external tools, execute domain-specific code, and perform metacognitive self-assessment for robust end-to-end task execution. We demonstrate three practical capabilities: a high-fidelity polymer property prediction and generative design pipeline, a fully automated multimodal workflow for biopolymer structure characterization, and a metacognitive agent framework that can monitor performance and improve execution strategies over time. On a held-out test set of 1,251 polymers, our PolyGNN agent achieves strong predictive accuracy, reaching R2 = 0.89 for glass-transition temperature (Tg ), R2 = 0.82 for tensile strength, R2 = 0.75 for elongation, and R2 = 0.91 for density. The framework also provides uncertainty estimates via multiagent consensus and scales with linear complexity to at least 10,000 polymers, enabling high-throughput screening at low computational cost. For a representative workload, the system completes inference in 16.3 s using about 2 GB of memory and 0.1 GPU hours, at an estimated cost of about $0.08. On a dedicated Tg benchmark, our approach attains R2 = 0.78, outperforming strong baselines including single-LLM prediction (R2 = 0.67), group-contribution methods (R2 = 0.71), and ChemCrow (R2 = 0.66). We further demonstrate metacognitive control in a polystyrene case study, where the system not only produces domain-level scientific outputs but continually monitors and optimizes its own behavior through tactical, strategic, and meta-strategic self-assessment.
Problem
Research questions and friction points this paper is trying to address.
polymer informatics
property prediction
generative design
high-throughput screening
multi-agent AI
Innovation
Methods, ideas, or system contributions that make the work stand out.
multi-agent AI
polymer informatics
metacognitive reasoning
generative design
high-throughput screening
🔎 Similar Papers
No similar papers found.
