About this Abstract |
| Meeting |
2025 TMS Annual Meeting & Exhibition
|
| Symposium
|
Thermodynamics and Kinetics of Alloys III
|
| Presentation Title |
Integration of Large-Language Model and CALPHAD for Alloy Design Hypothesis Generation |
| Author(s) |
Quanliang Liu, Hyunseok Oh |
| On-Site Speaker (Planned) |
Hyunseok Oh |
| Abstract Scope |
The increasing complexity of materials design requires advanced methods to manage and transfer extensive information across various domains. With in-context learning capabilities, large-language models (LLMs) provide a new opportunity for streamlining and synthesizing literature to generate new alloy design hypotheses. However, a challenge in applying LLMs for generating hypotheses is a knowledge gap, as LLMs’ text synthesis is often not constrained by physical rules unless creativity is sacrificed. In this research, we integrate CALPHAD to bridge the knowledge gap in LLM-generated hypotheses. The CALPHAD problems and parameters are suggested by LLMs, and the results provide iterative feedback to refine LLM-generated hypotheses. This approach, guided by CALPHAD, enhances the potential for disruptive discoveries of new materials while maintaining scientific and engineering plausibility. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Computational Materials Science & Engineering, High-Entropy Alloys, Iron and Steel |