Towards Edge General Intelligence: Knowledge Distillation for Mobile Agentic AI
Yuxuan Wu, Linghan Ma, Ruichen Zhang, Yinqiu Liu, Dusit Niyato, Fellow, IEEE · 2025
Read Towards Edge General Intelligence: Knowledge Distillation for Mobile Agentic AI by Yuxuan Wu, Linghan Ma, Ruichen Zhang, Yinqiu Liu, Dusit Niyato, Fellow, IEEE (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-2511-19947.txt.
Description
Yuxuan Wu and eleven co-authors survey how knowledge distillation can bring agentic AI to mobile devices. Edge general intelligence means agents acting on their own in changing, resource-poor settings, but advanced models strain mobile limits on computing, energy, and storage. Distillation, where a small model learns from a larger one, is their main answer.
The survey stresses methods built for wireless settings, such as channel-aware self-distillation, distillation of channel state feedback, and robust signal classification. It reviews model families suited to the edge, including Mamba and RWKV, and cross-architecture distillation. Applications span vision, speech, and multimodal tasks. The authors close with open problems and future directions for the field.
Catalog record
| CREATOR | Yuxuan Wu, Linghan Ma, Ruichen Zhang, Yinqiu Liu, Dusit Niyato, Fellow, IEEE |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 90 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2511-19947