Enhancing Long-Range Dependency with State Space Model and Kolmogorov-Arnold Networks for Aspect-Based Sentiment Analysis
Adamu Lawan, Juhua Pu, Haruna Yunusa, Aliyu Umar, Muhammad Lawan
TL;DR
This work tackles ABSA by addressing long-range aspect-opinion dependencies and the inefficiency of traditional attention. It introduces MambaForGCN, which fuses syntax-aware graphs (SynGCN) with a semantic MambaFormer module and a KAN-based gated fusion to capture both short- and long-range cues, including implicit opinions. Across three benchmark datasets, the approach outperforms several SOTA baselines, with additional gains when combined with BERT, highlighting the practical value of integrating syntactic structure, state-space models, and non-linear feature fusion. The results underscore the potential of non-linear, adaptive representations for robust ABSA, while noting generalization challenges to diverse real-world texts.
Abstract
Aspect-based Sentiment Analysis (ABSA) evaluates sentiments toward specific aspects of entities within the text. However, attention mechanisms and neural network models struggle with syntactic constraints. The quadratic complexity of attention mechanisms also limits their adoption for capturing long-range dependencies between aspect and opinion words in ABSA. This complexity can lead to the misinterpretation of irrelevant contextual words, restricting their effectiveness to short-range dependencies. To address the above problem, we present a novel approach to enhance long-range dependencies between aspect and opinion words in ABSA (MambaForGCN). This approach incorporates syntax-based Graph Convolutional Network (SynGCN) and MambaFormer (Mamba-Transformer) modules to encode input with dependency relations and semantic information. The Multihead Attention (MHA) and Selective State Space model (Mamba) blocks in the MambaFormer module serve as channels to enhance the model with short and long-range dependencies between aspect and opinion words. We also introduce the Kolmogorov-Arnold Networks (KANs) gated fusion, an adaptive feature representation system that integrates SynGCN and MambaFormer and captures non-linear, complex dependencies. Experimental results on three benchmark datasets demonstrate MambaForGCN's effectiveness, outperforming state-of-the-art (SOTA) baseline models.