Knowledge-informed graph attention networks enable defect-free alloy design for laser additive manufacturing
Jiabo Fu, Hao Yu, Chenchong Wang, Lingyu Wang, Zhongji Sun, Jinguo Li, Sybrand van der Zwaag, Dierk Raabe, Wei Xu · 2026
Read Knowledge-informed graph attention networks enable defect-free alloy design for laser additive manufacturing by Jiabo Fu, Hao Yu, Chenchong Wang, Lingyu Wang, Zhongji Sun, Jinguo Li, Sybrand van der Zwaag, Dierk Raabe, Wei Xu (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13612592.txt.
Description
Jiabo Fu and co-authors wrote a study from 2026 on designing printer-friendly metal alloys with a knowledge-guided form of machine learning. Laser additive manufacturing builds parts layer by layer, but new alloys usually demand either heavy physics modeling or huge data sets. Data for fresh alloys are scarce, and naive models overfit or propose nonsense. The team wired textbook metallurgy knowledge into a graph attention network and tracked how unsure the defect data are.
The framework balanced that built-in knowledge against the scatter in real defect counts. As its test case, it produced a new nickel superalloy free of cracks and pores, with strong laser printability, heat stability, and strength. Atom-level probing confirmed the guessed reason for the performance. The same approach then yielded a printable aluminum alloy. The authors offer the method as a faster route to next-generation alloys for printing.
Catalog record
| CREATOR | Jiabo Fu, Hao Yu, Chenchong Wang, Lingyu Wang, Zhongji Sun, Jinguo Li, Sybrand van der Zwaag, Dierk Raabe, Wei Xu |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 61 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |