A*STAR Career Development Fund (CDF) Day 2026 – Poster Award Winner
Through its Career Development Fund (CDF), A*STAR nurtures the next generation of scientific leaders by supporting promising early-career researchers in bold, interdisciplinary work. The annual A*STAR CDF Day celebrates these achievements by bringing together researchers to exchange ideas, foster collaborations, and showcase innovative research through poster presentations.
Congratulations to Dr Qian Hangwei, Scientist from A*STAR Centre for Frontier AI Research (A*STAR CFAR), on being named the A*STAR CDF Day 2026 Poster Award Winner for the project:
Building Material Intelligence: A Multimodal Pathway to Composite Material Design
Composite materials are widely used in aerospace, transportation, renewable energy, and advanced manufacturing due to their exceptional strength-to-weight ratio and highly tunable properties. However, designing next-generation composites remains a challenging and time-consuming process, requiring researchers to navigate enormous design spaces while balancing multiple performance objectives.
Dr Qian's research aims to establish Material Intelligence, a new AI paradigm that enables machines to understand, reason about, and design advanced materials through multimodal learning. Moving beyond traditional approaches based solely on numerical simulations or experimental data, the project integrates heterogeneous sources of information, including material microstructures, simulation results, physical properties, and scientific knowledge, into unified AI models capable of learning rich representations of materials.
Building upon this multimodal foundation, the research develops AI systems that support inverse material design, allowing scientists to specify desired material properties and automatically generate candidate material structures. By combining multimodal representation learning, scientific reasoning, and generative AI, the project seeks to accelerate the discovery of high-performance composite materials while reducing the cost and time associated with traditional trial-and-error approaches.
This work lays the foundation for next-generation AI systems capable of assisting scientific discovery across a broad range of disciplines. Looking ahead, the project envisions intelligent AI agents that can integrate knowledge from simulations, experiments, and scientific literature, collaborate with researchers, and continuously learn to accelerate innovation in materials science and beyond.
References
[1] Learning ordinality-aware multimodal representations for composite materials design. Nature Communications (2026).
https://www.nature.com/articles/s41467-026-75615-3
[2] Geometry-Aware OOD Generalisation for Composite Materials. KDD (2026). https://dl.acm.org/doi/10.1145/3770855.3819056