From large language models to AI agents in energy materials research: enabling...

Paper

Abstract

Fragmented knowledge and slow experimental iteration constrain the discovery of energy materials. We trace the evolution of artificial intelligence (AI) in materials science, from large language models as knowledge assistants to autonomous agents that can reason, plan, and use tools. We introduce a two-path framework to analyze this evolution, distinguishing architectural innovation (agent collaboration) from cognitive innovation (learning and representation). This framework synthesizes recent progress in AI-driven discovery, design, and automation. By examining challenges in reliability, interpretability, and physical grounding, we outline a roadmap toward physics-informed, human-AI systems for autonomous scientific discovery.

Analysis

Paper intelligence

Summary

This paper explores the evolution of AI in energy materials research, from knowledge assistants to autonomous agents, proposing a framework to analyze progress and outlining a roadmap for physics-informed, human-AI systems.

Key findings

AI is evolving from knowledge assistants to autonomous agents in materials research.
A two-path framework (architectural and cognitive innovation) synthesizes AI progress.
Challenges in reliability, interpretability, and physical grounding need addressing.

Methods

Tracing the evolution of AI in materials science.
Introducing a two-path framework (architectural and cognitive innovation).
Examining challenges in reliability, interpretability, and physical grounding.

Limitations

Reliability of AI systems in materials research.
Interpretability of AI models.
Physical grounding of AI-driven discoveries.

Applications

Energy materials discovery.
Energy materials design.
Automation of energy materials research.

Confidence100%

Discovery

Related papers

2026Crossref0 citations

Improving predictive modeling in materials science based on data augmentation approach

Mohamed Hamidi; Mohamed Loutou

Jul 10, 2026Journal of Materials Science: Materials in Engineering

2026Crossref0 citations

Transport Novelty Distance: A Distributional Metric for Evaluating Material Generative Models

Paul Hagemann; Simon Müller; Janine George; Philipp Benner

Jul 6, 2026Journal of Physics: Materials

2026Crossref0 citations

Machine Learning in Materials Science: Data-Driven Discovery and Functional Applications

Mihail Kolev

Jul 6, 2026Encyclopedia

2026Crossref0 citations

Matter - AI A Multimodal AI Framework for Accelerated Materials Discovery in Physical Sciences

Prudvi Saisaran Ponduru

Jun 14, 2026International Journal of Innovative Research in Engineering & Multidisciplinary Physical Sciences

2026Crossref0 citations

AI/ML Enabled High‐Throughput Design and Synthesis for Energetic Molecules

Wen Qian

May 19, 2026Materials Genome Engineering Advances

2026Crossref1 citations

Generative Design for Alloys: Harnessing Generative Models for Faster Discovery

Cheng Li; Yuehui Xian; Yumei Zhou; Xiangdong Ding; Jun Sun; Dezhen Xue

May 1, 2026Advanced Materials

Source record

Publication metadata

Authors: Tongao Yao; Junming Huang; Yujie Yan; Yang Yang; Ziye Wang; Xuqiang Shao; Zhengyang Gao; Weijie Yang
Published: Dec 19, 2025
Journal or venue: AI Agent
DOI: 10.20517/aiagent.2025.03
Paper type: journal article
Source: Crossref

DOI