Quantum-Inspired Spectral Geometry for Neural Operator Equivalence and Structured Pruning
Haijian Shao, Wei Liu, Xing Deng School of Computer, Jiangsu University of Science and Technology, Zhenjiang 212003, China Qi Liu School of Computer and Software, Nanjing University of Information Science and Technology, Nanjing 210044, China Yingtao Jiang Department of Electrical and Computer Engin · 2025
Read Quantum-Inspired Spectral Geometry for Neural Operator Equivalence and Structured Pruning by Haijian Shao, Wei Liu, Xing Deng School of Computer, Jiangsu University of Science and Technology, Zhenjiang 212003, China Qi Liu School of Computer and Software, Nanjing University of Information Science and Technology, Nanjing 210044, China Yingtao Jiang Department of Electrical and Computer Engin (2025) 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/arxiv-2512-00880.txt.
Description
Haijian Shao, Wei Liu, Xing Deng, Qi Liu, and Yingtao Jiang propose a quantum-inspired method for neural operators in a 2025 paper. Multimodal AI on home-grade hardware faces tight limits: mixed data types, real-time demands, and repeated operator blocks that waste capacity. Neural operators are the building blocks that transform data inside such models.
Their framework describes each operator by its singular value spectrum, a fingerprint of how it stretches data, placed on a geometric sphere. That view lets them test when two operators are equivalent and prune ones that repeat work. Leaner models could run better on modest devices.
Catalog record
| CREATOR | Haijian Shao, Wei Liu, Xing Deng School of Computer, Jiangsu University of Science and Technology, Zhenjiang 212003, China Qi Liu School of Computer and Software, Nanjing University of Information Science and Technology, Nanjing 210044, China Yingtao Jiang Department of Electrical and Computer Engin |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 22 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2512-00880