FastPOS: Language-Agnostic Scalable POS Tagging Framework Low-Resource Use Case
Fuente:
arXiv
Salvato in:
| Autori principali: | , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914174909546496 |
|---|---|
| author | Kafi, Md Abdullah Al Banshal, Sumit Kumar |
| author_facet | Kafi, Md Abdullah Al Banshal, Sumit Kumar |
| contents | This study proposes a language-agnostic transformer-based POS tagging framework designed for low-resource languages, using Bangla and Hindi as case studies. With only three lines of framework-specific code, the model was adapted from Bangla to Hindi, demonstrating effective portability with minimal modification. The framework achieves 96.85 percent and 97 percent token-level accuracy across POS categories in Bangla and Hindi while sustaining strong F1 scores despite dataset imbalance and linguistic overlap. A performance discrepancy in a specific POS category underscores ongoing challenges in dataset curation. The strong results stem from the underlying transformer architecture, which can be replaced with limited code adjustments. Its modular and open-source design enables rapid cross-lingual adaptation while reducing model design and tuning overhead, allowing researchers to focus on linguistic preprocessing and dataset refinement, which are essential for advancing NLP in underrepresented languages. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_00745 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | FastPOS: Language-Agnostic Scalable POS Tagging Framework Low-Resource Use Case Kafi, Md Abdullah Al Banshal, Sumit Kumar Computation and Language This study proposes a language-agnostic transformer-based POS tagging framework designed for low-resource languages, using Bangla and Hindi as case studies. With only three lines of framework-specific code, the model was adapted from Bangla to Hindi, demonstrating effective portability with minimal modification. The framework achieves 96.85 percent and 97 percent token-level accuracy across POS categories in Bangla and Hindi while sustaining strong F1 scores despite dataset imbalance and linguistic overlap. A performance discrepancy in a specific POS category underscores ongoing challenges in dataset curation. The strong results stem from the underlying transformer architecture, which can be replaced with limited code adjustments. Its modular and open-source design enables rapid cross-lingual adaptation while reducing model design and tuning overhead, allowing researchers to focus on linguistic preprocessing and dataset refinement, which are essential for advancing NLP in underrepresented languages. |
| title | FastPOS: Language-Agnostic Scalable POS Tagging Framework Low-Resource Use Case |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2512.00745 |