Sina Ahmadi

Sina Ahmadi, Milind Agarwal, Antonios Anastasopoulos · cmu

The Perso-Arabic scripts are a family of scripts that are widely adopted and used by various linguistic communities around the globe. Identifying various languages using such scripts is crucial to language technologies and challenging in low-resource setups. As such, this paper sheds light on the challenges of detecting languages using Perso-Arabic scripts, especially in bilingual communities where ``unconventional'' writing is practiced. To address this, we use a set of supervised techniques to classify sentences into their languages. Building on these, we also propose a hierarchical model that targets clusters of languages that are more often confused by the classifiers. Our experiment results indicate the effectiveness of our solutions.

Sina Ahmadi, Zahra Azin, Sara Belelli et al. · cmu

One of the major challenges that under-represented and endangered language communities face in language technology is the lack or paucity of language data. This is also the case of the Southern varieties of the Kurdish and Laki languages for which very limited resources are available with insubstantial progress in tools. To tackle this, we provide a few approaches that rely on the content of local news websites, a local radio station that broadcasts content in Southern Kurdish and fieldwork for Laki. In this paper, we describe some of the challenges of such under-represented languages, particularly in writing and standardization, and also, in retrieving sources of data and retro-digitizing handwritten content to create a corpus for Southern Kurdish and Laki. In addition, we study the task of language identification in light of the other variants of Kurdish and Zaza-Gorani languages.

Razhan Hameed, Sina Ahmadi, Fatemeh Daneshfar

Sentiment analysis is the process of identifying and extracting subjective information from text. Despite the advances to employ cross-lingual approaches in an automatic way, the implementation and evaluation of sentiment analysis systems require language-specific data to consider various sociocultural and linguistic peculiarities. In this paper, the collection and annotation of a dataset are described for sentiment analysis of Central Kurdish. We explore a few classical machine learning and neural network-based techniques for this task. Additionally, we employ an approach in transfer learning to leverage pretrained models for data augmentation. We demonstrate that data augmentation achieves a high F$_1$ score and accuracy despite the difficulty of the task.

Sina Ahmadi

The focus of this thesis is broadly on the alignment of lexicographical data, particularly dictionaries. In order to tackle some of the challenges in this field, two main tasks of word sense alignment and translation inference are addressed. The first task aims to find an optimal alignment given the sense definitions of a headword in two different monolingual dictionaries. This is a challenging task, especially due to differences in sense granularity, coverage and description in two resources. After describing the characteristics of various lexical semantic resources, we introduce a benchmark containing 17 datasets of 15 languages where monolingual word senses and definitions are manually annotated across different resources by experts. In the creation of the benchmark, lexicographers' knowledge is incorporated through the annotations where a semantic relation, namely exact, narrower, broader, related or none, is selected for each sense pair. This benchmark can be used for evaluation purposes of word-sense alignment systems. The performance of a few alignment techniques based on textual and non-textual semantic similarity detection and semantic relation induction is evaluated using the benchmark. Finally, we extend this work to translation inference where translation pairs are induced to generate bilingual lexicons in an unsupervised way using various approaches based on graph analysis. This task is of particular interest for the creation of lexicographical resources for less-resourced and under-represented languages and also, assists in increasing coverage of the existing resources. From a practical point of view, the techniques and methods that are developed in this thesis are implemented within a tool that can facilitate the alignment task.

Isaac Caswell, Elizabeth Nielsen, Jiaming Luo et al. · mit

We open-source SMOL (Set of Maximal Overall Leverage), a suite of training data to unlock machine translation for low-resource languages. SMOL has been translated into 124 (and growing) under-resourced languages (125 language pairs), including many for which there exist no previous public resources, for a total of 6.1M translated tokens. SMOL comprises two sub-datasets, each carefully chosen for maximum impact given its size: SMOLSENT, a set of sentences chosen for broad unique token coverage, and SMOLDOC, a document-level resource focusing on a broad topic coverage. They join the already released GATITOS for a trifecta of paragraph, sentence, and token-level content. We demonstrate that using SMOL to prompt or fine-tune Large Language Models yields robust chrF improvements. In addition to translation, we provide factuality ratings and rationales for all documents in SMOLDOC, yielding the first factuality datasets for most of these languages.

Sina Ahmadi, Daban Q. Jaff, Md Mahfuz Ibn Alam et al. · cmu

Kurdish, an Indo-European language spoken by over 30 million speakers, is considered a dialect continuum and known for its diversity in language varieties. Previous studies addressing language and speech technology for Kurdish handle it in a monolithic way as a macro-language, resulting in disparities for dialects and varieties for which there are few resources and tools available. In this paper, we take a step towards developing resources for language and speech technology for varieties of Central Kurdish, creating a corpus by transcribing movies and TV series as an alternative to fieldwork. Additionally, we report the performance of machine translation, automatic speech recognition, and language identification as downstream tasks evaluated on Central Kurdish varieties. Data and models are publicly available under an open license at https://github.com/sinaahmadi/CORDI.

Charlotte Model, Sina Ahmadi, Jannis Vamvas

The Romansh language has several regional varieties, called idioms, which sometimes have limited mutual intelligibility. Despite this linguistic diversity, there has been a lack of documented efforts to build a language identification (LID) system that can distinguish between these idioms. Since Romansh LID should also be able to recognize Rumantsch Grischun, a supra-regional variety that combines elements of several idioms, this makes for a novel and interesting classification problem. In this paper, we present a LID system for Romansh idioms based on an SVM approach. We evaluate our model on a newly curated benchmark across two domains and find that it reaches an average in-domain accuracy of 97%, enabling applications such as idiom-aware spell checking or machine translation. Our classifier is publicly available.

Aso Mahmudi, Sina Ahmadi, Kemal Kurniawan et al.

Collecting and annotating morphological data present significant challenges, requiring linguistic expertise, methodological rigour, and substantial resources. These barriers are particularly acute for low-resource languages and varieties. To accelerate this process, we introduce \texttt{CommonMorph}, a comprehensive platform that streamlines morphological data collection development through a three-tiered approach: expert linguistic definition, contributor elicitation, and community validation. The platform minimises manual work by incorporating active learning, annotation suggestions, and tools to import and adapt materials from related languages. It accommodates diverse morphological systems, including fusional, agglutinative, and root-and-pattern morphologies. Its open-source design and UniMorph-compatible outputs ensure accessibility and interoperability with NLP tools. Our platform is accessible at https://common-morph.com, offering a replicable model for preserving linguistic diversity through collaborative technology.

Jannis Vamvas, Ignacio Pérez Prat, Angela Heldstab et al.

Recent strategies for low-resource machine translation rely on LLMs to generate synthetic data from higher-resource languages. We find that this method fails for Romansh, because LLMs tend to confuse its 6 distinct language varieties. Our experiments show that instead, the direction of data augmentation should be aligned with the resource gradient between source and target language. This approach surpasses Gemini 3 Pro in the lowest-resource variety of Romansh by 23 BLEU. A human evaluation confirms that our experiments yield the first model that generates fluent translations in the individual Romansh varieties.

Sina Ahmadi

Spell checking and morphological analysis are two fundamental tasks in text and natural language processing and are addressed in the early stages of the development of language technology. Despite the previous efforts, there is no progress in open-source to create such tools for Sorani Kurdish, also known as Central Kurdish, as a less-resourced language. In this paper, we present our efforts in annotating a lexicon with morphosyntactic tags and also, extracting morphological rules of Sorani Kurdish to build a morphological analyzer, a stemmer and a spell-checking system using Hunspell. This implementation can be used for further developments in the field by researchers and also, be integrated into text editors under a publicly available license.

Verena Platzgummer, John McCrae, Sina Ahmadi

The design of Large Language Models and generative artificial intelligence has been shown to be "unfair" to less-spoken languages and to deepen the digital language divide. Critical sociolinguistic work has also argued that these technologies are not only made possible by prior socio-historical processes of linguistic standardisation, often grounded in European nationalist and colonial projects, but also exacerbate epistemologies of language as "monolithic, monolingual, syntactically standardized systems of meaning". In our paper, we draw on earlier work on the intersections of technology and language policy and bring our respective expertise in critical sociolinguistics and computational linguistics to bear on an interrogation of these arguments. We take two different complexes of non-standard linguistic varieties in our respective repertoires--South Tyrolean dialects, which are widely used in informal communication in South Tyrol, Italy, as well as varieties of Kurdish--as starting points to an interdisciplinary exploration of the intersections between GenAI and linguistic variation and standardisation. We discuss both how LLMs can be made to deal with nonstandard language from a technical perspective, and whether, when or how this can contribute to "democratic and decolonial digital and machine learning strategies", which has direct policy implications.

Mérilin Sousa Silva, Sina Ahmadi

Throughout language history, words are borrowed from one language to another and gradually become integrated into the recipient's lexicon. Speakers can often differentiate these loanwords from native vocabulary, particularly in bilingual communities where a dominant language continuously imposes lexical items on a minority language. This paper investigates whether pretrained language models, including large language models, possess similar capabilities for loanword identification. We evaluate multiple models across 10 languages. Despite explicit instructions and contextual information, our results show that models perform poorly in distinguishing loanwords from native ones. These findings corroborate previous evidence that modern NLP systems exhibit a bias toward loanwords rather than native equivalents. Our work has implications for developing NLP tools for minority languages and supporting language preservation in communities under lexical pressure from dominant languages.

Joel Niklaus, Jakob Merane, Luka Nenadic et al.

In Switzerland legal translation is uniquely important due to the country's four official languages and requirements for multilingual legal documentation. However, this process traditionally relies on professionals who must be both legal experts and skilled translators -- creating bottlenecks and impacting effective access to justice. To address this challenge, we introduce SwiLTra-Bench, a comprehensive multilingual benchmark of over 180K aligned Swiss legal translation pairs comprising laws, headnotes, and press releases across all Swiss languages along with English, designed to evaluate LLM-based translation systems. Our systematic evaluation reveals that frontier models achieve superior translation performance across all document types, while specialized translation systems excel specifically in laws but under-perform in headnotes. Through rigorous testing and human expert validation, we demonstrate that while fine-tuning open SLMs significantly improves their translation quality, they still lag behind the best zero-shot prompted frontier models such as Claude-3.5-Sonnet. Additionally, we present SwiLTra-Judge, a specialized LLM evaluation system that aligns best with human expert assessments.

Negar Foroutan, Clara Meister, Debjit Paul et al.

Tokenization is the first -- and often least scrutinized -- step of most NLP pipelines. Standard algorithms for learning tokenizers rely on frequency-based objectives, which favor languages dominant in the training data and consequently leave lower-resource languages with tokenizations that are disproportionately longer, morphologically implausible, or even riddled with <UNK> placeholders. This phenomenon ultimately amplifies computational and financial inequalities between users from different language backgrounds. To remedy this, we introduce Parity-aware Byte Pair Encoding (BPE), a variant of the widely-used BPE algorithm. At every merge step, Parity-aware BPE maximizes the compression gain of the currently worst-compressed language, trading a small amount of global compression for cross-lingual parity. We find empirically that Parity-aware BPE leads to more equitable token counts across languages, with negligible impact on global compression rate and no substantial effect on language-model performance in downstream tasks.

Md Mahfuz Ibn Alam, Sina Ahmadi, Antonios Anastasopoulos · cmu

The availability of parallel texts is crucial to the performance of machine translation models. However, most of the world's languages face the predominant challenge of data scarcity. In this paper, we propose strategies to synthesize parallel data relying on morpho-syntactic information and using bilingual lexicons along with a small amount of seed parallel data. Our methodology adheres to a realistic scenario backed by the small parallel seed data. It is linguistically informed, as it aims to create augmented data that is more likely to be grammatically correct. We analyze how our synthetic data can be combined with raw parallel data and demonstrate a consistent improvement in performance in our experiments on 14 languages (28 English <-> X pairs) ranging from well- to very low-resource ones. Our method leads to improvements even when using only five seed sentences and a bilingual lexicon.

Tyler A. Chang, Catherine Arnett, Abdelrahman Eldesokey et al. · uw

To date, there exist almost no culturally-specific evaluation benchmarks for large language models (LLMs) that cover a large number of languages and cultures. In this paper, we present Global PIQA, a participatory commonsense reasoning benchmark for over 100 languages, constructed by hand by 335 researchers from 65 countries around the world. The 116 language varieties in Global PIQA cover five continents, 14 language families, and 23 writing systems. In the non-parallel split of Global PIQA, over 50% of examples reference local foods, customs, traditions, or other culturally-specific elements. We find that state-of-the-art LLMs perform well on Global PIQA in aggregate, but they exhibit weaker performance in lower-resource languages (up to a 37% accuracy gap, despite random chance at 50%). Open models generally perform worse than proprietary models. Global PIQA highlights that in many languages and cultures, everyday knowledge remains an area for improvement, alongside more widely-discussed capabilities such as complex reasoning and expert knowledge. Beyond its uses for LLM evaluation, we hope that Global PIQA provides a glimpse into the wide diversity of cultures in which human language is embedded.

Md Mahfuz Ibn Alam, Sina Ahmadi, Antonios Anastasopoulos

Neural machine translation (NMT) systems exhibit limited robustness in handling source-side linguistic variations. Their performance tends to degrade when faced with even slight deviations in language usage, such as different domains or variations introduced by second-language speakers. It is intuitive to extend this observation to encompass dialectal variations as well, but the work allowing the community to evaluate MT systems on this dimension is limited. To alleviate this issue, we compile and release CODET, a contrastive dialectal benchmark encompassing 891 different variations from twelve different languages. We also quantitatively demonstrate the challenges large MT models face in effectively translating dialectal variants. All the data and code have been released.

Sina Ahmadi, Antonios Anastasopoulos

The wide accessibility of social media has provided linguistically under-represented communities with an extraordinary opportunity to create content in their native languages. This, however, comes with certain challenges in script normalization, particularly where the speakers of a language in a bilingual community rely on another script or orthography to write their native language. This paper addresses the problem of script normalization for several such languages that are mainly written in a Perso-Arabic script. Using synthetic data with various levels of noise and a transformer-based model, we demonstrate that the problem can be effectively remediated. We conduct a small-scale evaluation of real data as well. Our experiments indicate that script normalization is also beneficial to improve the performance of downstream tasks such as machine translation and language identification.

Sina Ahmadi

Sorani Kurdish, also known as Central Kurdish, has a complex morphology, particularly due to the patterns in which morphemes appear. Although several aspects of Kurdish morphology have been studied, such as pronominal endoclitics and Izafa constructions, Sorani Kurdish morphology has received trivial attention in computational linguistics. Moreover, some morphemes, such as the emphasis endoclitic =îş, and derivational morphemes have not been previously studied. To tackle the complex morphology of Sorani, we provide a thorough description of Sorani Kurdish morphological and morphophonological constructions in a formal way such that they can be used as finite-state transducers for morphological analysis and synthesis.

Sina Ahmadi, Mariam Masoud

Machine translation is the task of translating texts from one language to another using computers. It has been one of the major tasks in natural language processing and computational linguistics and has been motivating to facilitate human communication. Kurdish, an Indo-European language, has received little attention in this realm due to the language being less-resourced. Therefore, in this paper, we are addressing the main issues in creating a machine translation system for the Kurdish language, with a focus on the Sorani dialect. We describe the available scarce parallel data suitable for training a neural machine translation model for Sorani Kurdish-English translation. We also discuss some of the major challenges in Kurdish language translation and demonstrate how fundamental text processing tasks, such as tokenization, can improve translation performance.

Sina Ahmadi, Hossein Hassani, Daban Q. Jaff

Machine translation has been a major motivation of development in natural language processing. Despite the burgeoning achievements in creating more efficient machine translation systems thanks to deep learning methods, parallel corpora have remained indispensable for progress in the field. In an attempt to create parallel corpora for the Kurdish language, in this paper, we describe our approach in retrieving potentially-alignable news articles from multi-language websites and manually align them across dialects and languages based on lexical similarity and transliteration of scripts. We present a corpus containing 12,327 translation pairs in the two major dialects of Kurdish, Sorani and Kurmanji. We also provide 1,797 and 650 translation pairs in English-Kurmanji and English-Sorani. The corpus is publicly available under the CC BY-NC-SA 4.0 license.

Sina Ahmadi, Hossein Hassani

Morphological analysis is the study of the formation and structure of words. It plays a crucial role in various tasks in Natural Language Processing (NLP) and Computational Linguistics (CL) such as machine translation and text and speech generation. Kurdish is a less-resourced multi-dialect Indo-European language with highly inflectional morphology. In this paper, as the first attempt of its kind, the morphology of the Kurdish language (Sorani dialect) is described from a computational point of view. We extract morphological rules which are transformed into finite-state transducers for generating and analyzing words. The result of this research assists in conducting studies on language generation for Kurdish and enhances the Information Retrieval (IR) capacity for the language while leveraging the Kurdish NLP and CL into a more advanced computational level.

Roshna Omer Abdulrahman, Hossein Hassani, Sina Ahmadi

Kurdish is a less-resourced language consisting of different dialects written in various scripts. Approximately 30 million people in different countries speak the language. The lack of corpora is one of the main obstacles in Kurdish language processing. In this paper, we present KTC-the Kurdish Textbooks Corpus, which is composed of 31 K-12 textbooks in Sorani dialect. The corpus is normalized and categorized into 12 educational subjects containing 693,800 tokens (110,297 types). Our resource is publicly available for non-commercial use under the CC BY-NC-SA 4.0 license.

Sina Ahmadi

In this article, we present a rule-based approach for transliterating two mostly used orthographies in Sorani Kurdish. Our work consists of detecting a character in a word by removing the possible ambiguities and mapping it into the target orthography. We describe different challenges in Kurdish text mining and propose novel ideas concerning the transliteration task for Sorani Kurdish. Our transliteration system, named Wergor, achieves 82.79% overall precision and more than 99% in detecting the double-usage characters. We also present a manually transliterated corpus for Kurdish.

Sina Ahmadi

Morphological declension, which aims to inflect nouns to indicate number, case and gender, is an important task in natural language processing (NLP). This research proposal seeks to address the degree to which Recurrent Neural Networks (RNNs) are efficient in learning to decline noun cases. Given the challenge of data sparsity in processing morphologically rich languages and also, the flexibility of sentence structures in such languages, we believe that modeling morphological dependencies can improve the performance of neural network models. It is suggested to carry out various experiments to understand the interpretable features that may lead to a better generalization of the learned models on cross-lingual tasks.

Shahin Salavati, Sina Ahmadi

The present paper aims at presenting a lemmatization and a word-level error correction system for Sorani Kurdish. We propose a hybrid approach based on the morphological rules and a n-gram language model. We have called our lemmatization and error correction systems Peyv and Rênûs respectively, which are the first tools presented for Sorani Kurdish to the best of our knowledge. The Peyv lemmatizer has shown 86.7% accuracy. As for Rênûs, using a lexicon, we have obtained 96.4% accuracy while without a lexicon, the correction system has 87% accuracy. As two fundamental text processing tools, these tools can pave the way for further researches on more natural language processing applications for Sorani Kurdish.

Sina Ahmadi

Automatic spelling and grammatical correction systems are one of the most widely used tools within natural language applications. In this thesis, we assume the task of error correction as a type of monolingual machine translation where the source sentence is potentially erroneous and the target sentence should be the corrected form of the input. Our main focus in this project is building neural network models for the task of error correction. In particular, we investigate sequence-to-sequence and attention-based models which have recently shown a higher performance than the state-of-the-art of many language processing problems. We demonstrate that neural machine translation models can be successfully applied to the task of error correction. While the experiments of this research are performed on an Arabic corpus, our methods in this thesis can be easily applied to any language.