Citation

Lim, Jia Min and Lee, Chin Poo and Lim, Kian Ming and Ong, Hway Boon (2026) Adaptive Spatial Token Gating for Efficient Visual Inference. In: 3rd IEEE International Conference on Energy and Electrical Engineering, EEE 2026, 26 June 2026 - 27 June 2026, Nanchang.

Text 3.pdf - Published Version Restricted to Repository staff only Download (689kB)

Abstract

Dense visual backbones aggregate all spatial tokens under the assumption of uniform contribution, although discriminative evidence is often spatially localized. This work introduces Adaptive Spatial Token Gating (ASTG), a lightweight and architecture-agnostic module that performs deterministic top-

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	Deep Learning, Plant Disease Recognition, Image Classification
Subjects:	Q Science > QA Mathematics > QA71-90 Instruments and machines > QA75.5-76.95 Electronic computers. Computer science
Divisions:	Faculty of Information Science and Technology (FIST)
Depositing User:	Ms Rosnani Abd Wahab
Date Deposited:	28 Oct 2026 04:12
Last Modified:	28 Oct 2026 04:12
URII:	http://shdl.mmu.edu.my/id/eprint/16936

Downloads

Edit (login required)