用KAN网络提升缅甸语新闻分类效果，比传统方法更高效准确。
Enhancing Burmese News Classification with Kolmogorov-Arnold Network Head Fine-tuning
- 用KAN替代MLP做分类头，利用其可学习非线性增强表达能力
- fastText+EfficientKAN达0.928 F1，优于其他模型
- 适合低资源语言任务，尤其关注速度与精度平衡的场景
在缅甸语等低资源语言中，分类任务通常仅微调最终分类层，冻结预训练编码器权重。尽管多层感知机（MLPs）被广泛使用，但其固定非线性限制了表达能力并增加计算开销。本文探索了柯尔莫戈罗夫-阿诺德网络（KANs）作为替代分类头，评估了基于傅里叶的FourierKAN、基于样条的EfficientKAN和基于网格的FasterKAN，在TF-IDF、fastText及多语言Transformer（mBERT、Distil-mBERT）等多种嵌入表示上的表现。实验结果表明，基于KAN的分类头在性能上可媲美或超越MLPs。fastText结合EfficientKAN取得最高F1分数（0.928），FasterKAN在速度与精度间表现最佳。在Transformer嵌入上，EfficientKAN与mBERT配合达到0.917 F1，与MLP相当或略优。这些发现表明，KAN是低资源语言分类中表达能力强、效率高的替代方案。
原文摘要 · Abstract (English)
In low-resource languages like Burmese, classification tasks often fine-tune only the final classification layer, keeping pre-trained encoder weights frozen. While Multi-Layer Perceptrons (MLPs) are commonly used, their fixed non-linearity can limit expressiveness and increase computational cost. This work explores Kolmogorov-Arnold Networks (KANs) as alternative classification heads, evaluating Fourier-based FourierKAN, Spline-based EfficientKAN, and Grid-based FasterKAN-across diverse embeddings including TF-IDF, fastText, and multilingual transformers (mBERT, Distil-mBERT). Experimental results show that KAN-based heads are competitive with or superior to MLPs. EfficientKAN with fastText achieved the highest F1-score (0.928), while FasterKAN offered the best trade-off between speed and accuracy. On transformer embeddings, EfficientKAN matched or slightly outperformed MLPs with mBERT (0.917 F1). These findings highlight KANs as expressive, efficient alternatives to MLPs for low-resource language classification.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。