Cosine Similarity Distillation Vision Mixture-of-Experts for Intelligent Housing-Dimensional Urban Physical Examinations
Kun Zhao, Helei Ren, Wenbin He, Yuhong Zhao, Jinming Jiang, Wanxiang Yao, Weijun Gao, Qichao Ban · 2026
Read Cosine Similarity Distillation Vision Mixture-of-Experts for Intelligent Housing-Dimensional Urban Physical Examinations by Kun Zhao, Helei Ren, Wenbin He, Yuhong Zhao, Jinming Jiang, Wanxiang Yao, Weijun Gao, Qichao Ban (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-13258879.txt.
Description
Kun Zhao and colleagues inspect aging housing conditions from photos with a mixture of experts vision model in a 2026 study. A new labeled image dataset supports graded checks of building problems in older communities. Their model routes image regions to specialist subnetworks using a similarity based rule.
A graph method converted pretrained weights into coherent expert starting points. Special losses handled both broad categories and rare fine grained problems. Accuracy rose over a small standard vision baseline and over prior expert models. Computing costs fell at the same time. City renewal teams gain a faster screening aid for large housing inspection workloads.
Catalog record
| CREATOR | Kun Zhao, Helei Ren, Wenbin He, Yuhong Zhao, Jinming Jiang, Wanxiang Yao, Weijun Gao, Qichao Ban |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 60 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |