Symmetric Affix–Context Co-Attention: A Dual-Gating Framework for Robust POS Tagging in Low-Resource MRLs
Symmetry · 18.09.2025 · 17 (9) · 1561
DOI: 10.3390/sym17091561 ↗Kurz gefasst
Fachartikel in der Zeitschrift Symmetry, online erschienen am 18.09.2025. Verfasst von Yuan Qi und 2 weiteren Autor:innen. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Natural Science Foundation of Xinjiang Uyghur Autonomous Region, ational Natural Science Foundation of China Youth Program, Ministry of Education of Humanities and Social Science Project.
Abstract (Originalsprache)
Part-of-speech (POS) tagging in low-resource, morphologically rich languages (LRLs/MRLs) remains challenging due to extensive affixation, high out-of-vocabulary (OOV) rates, and pervasive polysemy. We propose MRL-POS, a unified Transformer-CRF framework that dynamically selects informative affix features and integrates them with deep contextual embeddings via a novel dual-gating co-attention mechanism. First, a Dynamic Affix Selector adaptively adjusts n-gram ranges and frequency thresholds based on word length to ensure high-precision affix segmentation. Second, the Affix–Context Co-Attention Module employs two gating functions that conditionally amplify contextual dimensions with affix cues and vice versa, enabling robust disambiguation of complex and ambiguous forms. Third, Layer-Wise Attention Pooling aggregates multi-layer XLM-RoBERTa representations, emphasizing those most relevant for morphological and syntactic tagging. Evaluations on Uyghur, Kyrgyz, and Uzbek show that MRL-POS achieves an average F1 of 84.10%, OOV accuracy of 84.24%, and Poly-F1 of 72.14%, outperforming strong baselines by up to 8 F1 points. By explicitly modeling the symmetry between morphological affix cues and sentence-level context through a dual-gating co-attention mechanism, MRL-POS achieves a balanced fusion that both preserves local structure and captures global dependencies. Interpretability analyses confirm that 89.1% of the selected affixes align with linguistic expectations. This symmetric design not only enhances robustness in low-resource and agglutinative settings but also offers a general paradigm for symmetry-aware sequence labeling tasks.
Förderung
- Natural Science Foundation of Xinjiang Uyghur Autonomous Region
- ational Natural Science Foundation of China Youth Program
- Ministry of Education of Humanities and Social Science Project
Zitieren
APA
Qi, Y., Ali, S., & Murat, A. (2025). Symmetric Affix–Context Co-Attention: A Dual-Gating Framework for Robust POS Tagging in Low-Resource MRLs. Symmetry, 17(9), 1561. https://doi.org/10.3390/sym17091561
Vancouver
Qi Y, Ali S, Murat A. Symmetric Affix–Context Co-Attention: A Dual-Gating Framework for Robust POS Tagging in Low-Resource MRLs. Symmetry. 2025;17(9):1561. doi:10.3390/sym17091561
BibTeX
@article{Qi2025,
author = {Qi, Yuan and Ali, Samat and Murat, Alim},
title = {Symmetric Affix–Context Co-Attention: A Dual-Gating Framework for Robust POS Tagging in Low-Resource MRLs},
journal = {Symmetry},
year = {2025},
volume = {17},
number = {9},
pages = {1561},
doi = {10.3390/sym17091561}
}
EUVIMED