Human-Constrained AI-Assisted Identification of Defect-Topology Relations for Mechanical Degradation in Porous Solids: A Discrete-Element Study
Yi Zhang, Yifei Liu, Heping Xie · 2026
Read Human-Constrained AI-Assisted Identification of Defect-Topology Relations for Mechanical Degradation in Porous Solids: A Discrete-Element Study by Yi Zhang, Yifei Liu, Heping Xie (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-13567631.txt.
Description
Zhang, Liu, and Xie let AI propose formulas for how porous solids weaken in this study from twenty twenty-six. Porosity counts missing material but says nothing about broken load paths. The team simulated over ten thousand bonded-particle specimens across five defect families. Human investigators fixed the rules: allowed inputs, physical priors, and a strict size cap.
Language models suggested descriptor ideas, which the authors curated into a measurement bench. Simple fitted formulas with six descriptors beat a three-dimensional neural network on the test bench. On fresh specimens the formulas predicted stiffness decay better than strength decay. The authors float a two-stage failure idea from that gap. All evidence is numerical, pending lab validation.
Catalog record
| CREATOR | Yi Zhang, Yifei Liu, Heping Xie |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 42 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |