LKAFormer: A Lightweight Kolmogorov-Arnold Transformer Model for Image Semantic Segmentation
ACM Transactions on Intelligent Systems and Technology · 30.06.2026 · 17 (3) · 1-24
DOI: 10.1145/3759254 ↗Kurz gefasst
Fachartikel in der Zeitschrift ACM Transactions on Intelligent Systems and Technology, online erschienen am 20.03.2026. Verfasst von Shoulin Yin und 8 weiteren Autor:innen. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 55 Literaturangaben. Gefördert von: Outstanding Young Science and Technology Talent in Dalian, Peak Climbing Plan Project of Dalian Municipal Central Hospital, Hainan Provincial Natural Science Foundation of China.
Abstract (Originalsprache)
Transformer-based semantic segmentation methods have demonstrated outstanding performance by leveraging global self-attention to effectively capture long-range dependence. However, there still exist two issues in existing works: (1) Most of them utilize the full-rank weight matrix to support the self-attention mechanism and feed-forward network in modelling long-range dependence between patches/pixels, resulting in a high computational cost during both training and inference. (2) Most of them ignore information interactions between high-level semantics and low-level structures during the image resolution recovery, which leads to the performance degradation in segmenting objects with complex boundaries. To tackle these challenges, a lightweight Kolmogorov-Arnold Transformer model (LKAFormer) is proposed for the image semantic segmentation, containing a two-stream lightweight Transformer encoder and a graph feature pyramid aggregation KAN-decoder. The former constructs a hierarchical feature cross-scale fusion pipeline to obtain sufficient semantics containing comprehensive multi-scale information via setting coarse-grained and fine-grained streams with different-size patches of images. In that pipeline, feature lightweight focusing modules model complex and long-range dependence across patches/pixels to refine image semantics with less computational costs by lightweight multi-head self-attention and lightweight feed-forward network designs. The latter leverages the learnable nonlinear transformation mechanism of the Kolmogorov-Arnold Transformer architecture to adaptively capture spatial structure dependence of distinct sub-regions of images. And then, it jointly performs the intra-scale graph fusion and cross-scale graph fusion during the image resolution recovery to enhance information interactions between high-level semantics and low-level structures, which achieves the robust boundary localization and texture refinement of segmentation objects. Finally, plentiful experiments are conducted on three challenging datasets, and the results show LKAFormer sets a new baseline in the image segmentation task in comparison with 11 methods.
Förderung
- Outstanding Young Science and Technology Talent in Dalian
- Peak Climbing Plan Project of Dalian Municipal Central Hospital
- Hainan Provincial Natural Science Foundation of China
- National Natural Science Foundation of China
- Central Guidance for Local Science and Technology Development Fund
- Liaoning Provincial Science and Technology Plan
- Fundamental Research Funds for the Central Universities
Zitieren
APA
Yin, S., Wang, L., Chen, T., Huang, H., Gao, J., Zhang, J., Liu, M., Li, P., & Xu, C. (2026). LKAFormer: A Lightweight Kolmogorov-Arnold Transformer Model for Image Semantic Segmentation. ACM Transactions on Intelligent Systems and Technology, 17(3), 1-24. https://doi.org/10.1145/3759254
Vancouver
Yin S, Wang L, Chen T, Huang H, Gao J, Zhang J, et al. LKAFormer: A Lightweight Kolmogorov-Arnold Transformer Model for Image Semantic Segmentation. ACM Transactions on Intelligent Systems and Technology. 2026;17(3):1-24. doi:10.1145/3759254
BibTeX
@article{Yin2026,
author = {Yin, Shoulin and Wang, Liguo and Chen, Tao and Huang, Huafei and Gao, Jing and Zhang, Jianing and Liu, Meng and Li, Peng and Xu, Chengpei},
title = {LKAFormer: A Lightweight Kolmogorov-Arnold Transformer Model for Image Semantic Segmentation},
journal = {ACM Transactions on Intelligent Systems and Technology},
year = {2026},
volume = {17},
number = {3},
pages = {1--24},
doi = {10.1145/3759254}
}
EUVIMED