Neural POS Tagging for the Low-Resource Saraiki

Usman Nawaz; Muhammad Junaid Iqbal; Muhammad Asaf; Mariam Munsif Mir; Faisal Rehman; Tahir Alyas; Muhammad Hasnain

cs.CLOct 14, 2026

Neural POS Tagging for the Low-Resource Saraiki

Authors: Usman Nawaz, Muhammad Junaid Iqbal, Muhammad Asaf, Mariam Munsif Mir, Faisal Rehman, Tahir Alyas, Muhammad Hasnain

Organizations: Independent Researcher, Palermo, Italy · University of Rome Tor Vergata, Rome, Italy · DIMES, University of Calabria, Rende, Italy · School of ICT, Griffith University, Nathan, QLD 4111, Australia ·

Abstract

Part-of-speech tagging is a fundamental sequence labeling task and an important component for many downstream natural language processing applications. However, Saraiki, written from right to left in a Perso-Arabic script, remains an under-resourced Indo-Aryan language, despite being spoken as a mother tongue by 28,849,579 people (approximately 28.9 million) according to Pakistan 7th Population and Housing Census 2023. This paper presents a manually annotated Saraiki Universal POS tagging dataset developed from plain-text. We annotated the text with 17 POS categories. We evaluate three pretrained transformer encoders under the same fine-tuning setup: XLM-RoBERTa base, Urdu RoBERTa small, and multilingual DistilBERT. Experimental results over five random seeds show that XLM-RoBERTa base achieves the best performance, with 87.29% +/- 0.21% development accuracy and 88.60% +/- 0.21% test accuracy. Urdu RoBERTa small and multilingual DistilBERT also provide competitive results, suggesting that multilingual and related-script pretrained encoders are effective for Saraiki UPOS tagging. The dataset, evaluation setup, and baseline results introduced in this work provide a benchmark for future research on Saraiki NLP and low-resource sequence labeling.

Figures & tables

Dataset Split	Sentences	Tokens
Training	1,666	28,377
Development	208	3,636
Test	209	3,439
Total	2,083	35,452

Table 1: Dataset statistics for the training, development, and test splits.

Tag	Train	Dev	Test	Total
ADJ	1,430	205	153	1,788
ADP	4,235	544	526	5,305
ADV	267	44	32	343
AUX	1,746	208	204	2,158
CCONJ	856	127	129	1,112
DET	73	12	8	93

Table 2: Number of tokens per UPOS tag in each data split.

Hyperparameter	Value
Maximum sequence length	256
Batch size	16
Epochs	10
Learning rate	$2\times 10^{-5}$
Weight decay	0.01
Dropout	0.10

Table 3: Hyperparameter settings for fine-tuning.

Model	Dev	Test
XLM-RoBERTa base	87.29 $\pm$ 0.21	88.60 $\pm$ 0.21
Urdu RoBERTa small	85.11 $\pm$ 0.14	86.85 $\pm$ 0.21
Multilingual DistilBERT	84.44 $\pm$ 0.32	86.03 $\pm$ 0.30

Table 4: UPOS accuracy (%) over five random seeds, reported as mean

\pm

sample standard deviation.

UPOS	Precision	Recall	F1	Support
ADJ	0.6826	0.7451	0.7125	153
ADP	0.9616	0.9525	0.9570	526
ADV	0.7083	0.5312	0.6071	32
AUX	0.8883	0.8971	0.8927	204
CCONJ	0.8603	0.9070	0.8830	129
DET	0.4615	0.7500	0.5714	8

Table 5: Per-class test results for XLM-RoBERTa (seed 46). Support is the number of gold test tokens.

Gold	AUX	NUM	NOUN	CCONJ	VERB	ADP	ADJ	PART	PROPN	PRON	SCONJ	ADV	DET
AUX	183	0	1	0	19	0	0	1	0	0	0	0	0
NUM	0	39	3	0	0	0	0	0	0	0	0	2	0
NOUN	4	4	911	1	23	7	30	0	53	11	4	1	1
CCONJ	0	0	0	117	0	10	0	0	0	0	2	0	0
VERB	13	0	26	0	178	0	5	1	1	3	0	0	1
ADP	0	1	5	17	1	501	1	0	0	0	0	0	0

Table 6: Test-set UPOS confusion matrix for XLM-RoBERTa (seed 46). Rows are gold tags; columns are predicted tags.

Morphosyntactic Tagging

Morphosyntactic tagging assigns words parts of speech and grammatical features such as case, number, gender, and tense.

See all 23 papers

Benchmarking POS Tagging for the Tajik Language: A Comparative Study of Neural Architectures on the TajPersParallel Corpus
May 6, 2026Mullosharaf K. ArabovMultilingual Language Model Evaluation Morphosyntactic Tagging
Institute of Computational Mathematics and Information Technologies Kazan Federal University Kazan, Russia
L3Cube-MahaPOS: A Marathi Part-of-Speech Tagging Dataset and BERT Models
Jun 23, 2026Hariom Ingle, Ronit Ghode, Ishwari Gondkar +2Morphosyntactic Tagging Low-Resource Language Processing
Department of Information Technology, PICT, Pune, India · L3 Cube Labs, Pune, India · Indian Institute of Technology Madras, Chennai, India
Under-resourced studies of under-resourced languages: lemmatization and POS-tagging with LLM annotators for historical Armenian, Georgian, Greek and Syriac
Feb 17, 2026Chahan Vidal-Gorène, Bastien Kindt, Florian CafieroLow-Resource Language Processing Morphosyntactic Tagging
CJM, LIPN · LIPN, CNRS UMR 7030, France · École nationale des chartes, Université Paris-Sciences-et-Lettres (PSL), France +3

May 6, 2026J/K move · Enter open · S save

Benchmarking POS Tagging for the Tajik Language: A Comparative Study of Neural Architectures on the TajPersParallel Corpus

Mullosharaf K. Arabov

Institute of Computational Mathematics and Information Technologies Kazan Federal University Kazan, Russia

Multilingual Language Model Evaluation Morphosyntactic Tagging Low-Resource Language Processing

Open paper AskPDF arXiv

This paper presents the first benchmark for the task of automatic part-of-speech (POS) tagging for the Tajik language. Despite the existence of multilingual language models demonstrating high effectiveness for many of the world's languages, their capacity for grammatical analysis of Tajik has remained unexplored until now. The aim of this study is to fill this gap through a systematic comparison of classical neural network architectures and modern multilingual transformers. Experiments were conducted on the TajPersParallel corpus, a parallel lexical resource comprising approximately 44,000 dictionary entries. Due to the absence of full-fledged example sentences in the current version of the corpus, the task was performed at the level of isolated lexical units, representing a challenging case of context-independent classification. The study compares the following architectures: a recurrent BiLSTM-CRF model, as well as multilingual models XLM-RoBERTa (large), mBERT, ParsBERT (Persian), and ruBERT (Russian), adapted using the parameter-efficient fine-tuning method LoRA. The testing results showed that the best performance is achieved by the mBERT + LoRA model (macro F1-score = 0.11, weighted F1-score = 0.62). It was established that in the absence of syntactic context, all models experience significant difficulty in resolving morphological ambiguity, successfully classifying primarily high-frequency classes ("noun," "adjective") while demonstrating zero effectiveness for rare function words. Zero-shot evaluation revealed the greatest typological proximity of Tajik to Persian (ParsBERT) and Russian (ruBERT). The obtained results form a foundation for further research and development in the field of automatic processing of the Tajik language.

Abstract

Morphosyntactic Tagging

Explore similar work

Benchmarking POS Tagging for the Tajik Language: A Comparative Study of Neural Architectures on the TajPersParallel Corpus

L3Cube-MahaPOS: A Marathi Part-of-Speech Tagging Dataset and BERT Models

Under-resourced studies of under-resourced languages: lemmatization and POS-tagging with LLM annotators for historical Armenian, Georgian, Greek and Syriac