Sep 2026· International Journal of Informatics and Communication Technology (IJ-ICT)· 0 citations· 59 references
TL;DR
This paper analyses various recent state-of-the-art variants of large language models (LLMs) and neural machine translation (NMT) for Indian languages in comparison to statistical machine translation (SMT) and tackles key questions, such as idiomatic expressions, morphologically complex grammar or the scarceness of parallel corpora.
Abstract
Significant advances have been achieved in machine translation (MT) in recent times, particularly state of the art (SOTA) models for languages like English and Indian having distinct grammatical structures and limited monolingual training data. This paper analyses various recent state-of-the-art variants of large language models (LLMs) and neural machine translation (NMT) for Indian languages in comparison to statistical machine translation (SMT). It tackles key questions, such as idiomatic expressions, morphologically complex grammar or the scarceness of parallel corpora. Furthermore, it studies bytewise BPE, compares translation models in terms of BLEU scores using separate and shared-vocabulary representation with copy actions between the BPE translations, and analyses how multitask learning (Caruana (1997)) and attention mechanisms can contribute to the quality of translation. In summary, it provides directions for future work by suggesting new avenues of research including better curated datasets, more efficient approaches for lowresource languages and culturally aware translations.
Machine translation (MT) technologies are currently undergoing a paradigm shift, transitioning
from specialized Neural Machine Translation (NMT) frameworks to the broader capabilities of
Large Language Models (LLMs). This paper examines the current standing of the Croatian language
within this technological evolution.
While bilingual NMT models often exhibit high precision, multilingual NMT leverage transfer
learning to enhance performance for low–resource language pairs, but with lower performance for
high–resource ones. Conversely, LLMs—whether general–purpose or fine–tuned for translation—
offer superior multilingual proficiency and context awareness. Unlike NMT, LLMs can process extended discourse, such as full paragraphs or documents, leading to significant improvements in
coreference resolution and gender agreement. Despite the substantial computational requirements
of LLMs, recent optimization techniques allow for smaller, more efficient versions that maintain
high output quality.
This study evaluates the performance of various NMT and LLM architectures specifically for
Croatian from/to English and Spanish using several automatic quality evaluation metrics. The findings demonstrate that open–source models can achieve, and occasionally surpass, the quality of
Google Translate, a widely used commercial NMT system. Furthermore, while our evaluation focuses on this specific language triad, the multilingual nature of the analysed systems suggests that
open–source models provide high–quality translation capabilities for Croatian across dozens, if not
hundreds, of language pairs.
Antoni Oliver, Sergi Álvarez–Vidal· Suvremena Lingvistika· 1 citation
India's linguistic landscape, comprising more than twenty scheduled languages and hundreds of additional dialects spanning multiple language families, presents a distinctive and severe challenge for machine translation (MT) systems predominantly developed and benchmarked on high-resource, Indo-European languages. This paper reviews the evolution of AI-driven natural language processing (NLP) approaches to Indian vernacular languages, tracing the shift from rule-based and statistical syntactic methods toward transformer-based semantic representation learning. The review synthesizes the transformer and multilingual pretraining literature, corpus-development efforts specific to Indian languages, and the growing evidence base on cross-lingual transfer and low-resource neural machine translation (NMT). Particular attention is given to the structural and morphological divergence between Indian languages and the English-centric architectures on which most large language models are trained, and to recent large-scale parallel-corpus and translation-model initiatives targeting this gap directly. Comparative tables summarize corpus scale, language coverage, and reported translation-quality metrics across the reviewed systems. The paper concludes that dedicated multilingual pretraining and large-scale parallel-corpus construction, rather than generic multilingual scaling alone, are the primary drivers of translation-quality gains for Indian vernacular languages, and identifies dialectal and code-mixed language coverage as the central future research prospect.
Dr. R. Sugunthakunthalambigai, Dr. Mallanna Biradar, Dr. Joyir Siram et al.· Anusandhanvallari· 0 citations
Neural machine translation (NMT) systems are widely used, but their performance remains strongly dependent on the availability of large-scale digital corpora, making translation for low-resource languages a persistent challenge. In parallel, large language models (LLMs) have recently emerged as a promising paradigm for multilingual text generation and translation; however, their behavior in low-resource settings remains largely underexplored. The challenge becomes even more acute for historical languages. Chagatai, a historical Turkic literary language of Central Asia with no native speakers, unstable orthography, and parallel data, represents an extreme case of such a condition. This study investigates whether transliteration significantly affects translation performance and how LLM-based and NMT-based systems compare under an extremely low-resource setting. To address these questions, we evaluated four source-text configurations (original Arabic script, expert manual transliteration, LLM-based transliteration, and rule-based Uroman transliteration) for translation into six target languages: Kazakh, English, Uzbek, Uyghur, Turkish, Russian, and Arabic. The results show that manual transliteration consistently yields the best translation performance, while noisy automatic romanization reduces these gains. For model comparison, GPT-4o was assessed alongside two fine-tuned NMT baselines, NLLB and TranslateGemma. The findings further show that LLM-based translation can be competitive with, and in some settings outperform, fine-tuned NMT systems, although this advantage comes with lower interpretability. Overall, these findings show that, for extremely low-resource historical languages written in non-Latin scripts, source-side representation is a decisive factor and may be as important as the choice of translation model itself.
A. Mansurova, Meruert Bekmukhamedova, Bekarys Baibolat et al.· Electronics· 0 citations
Machine Translation has become one of the major application areas of Artificial Intelligence (AI) and Natural Language Processing (NLP), especially in multilingual countries like India. Although recent Neural Machine Translation systems have shown good performance for several language pairs, translation quality is still inconsistent for many Indian languages because of linguistic and structural differences between English and Indian language families. Most Indian languages are morphologically rich and contain flexible word order, complex agreement patterns, compound constructions, and context-dependent grammatical forms. Because of this, direct translation from English often produces structurally incorrect or semantically weak output. In many existing systems, the source sentence is passed to the translation model without sufficient linguistic analysis. As a result, ambiguity present in the source text propagates further during translation. This work focuses on the importance of linguistic enrichment before the translation stage. The proposed framework, named Unified Linguistic-Aware Pre-Parsing Framework, introduces a coordinated pre-processing layer for English-to-Indian Machine Translation (MT). A key contribution of this research is the development of a novel linguistically enriched intermediate representation that extends beyond conventional text normalization. By transforming noisy input text into linguistically enriched translation-ready representation, the proposed approach facilitates effective knowledge transfer to machine translation models, leading to improve contextual adequacy, linguistic fidelity, and overall translation performance. The framework combines multiple linguistic processing stages including POS tagging, NE detection, clause boundary analysis, contextual token handling, syntactic structure preparation, and morphology-related processing. Instead of executing these modules independently, the proposed system allows interaction between lexical, syntactic, and morphological information during analysis. This helps reduce structural ambiguity and improves sentence-level interpretation before translation begins. The need for such a framework becomes more relevant in the context of Indian languages where morphology and grammatical relations carry significant semantic information. This framework is especially relevant for Indian languages, where semantic information is often encoded through morphological variations and grammatical dependencies. The proposed framework can be effectively integrated with both conventional machine translation architectures and modern large language models. The overall study highlights how classical linguistic analysis can still play an important role in improving multilingual AI systems for Indian languages.
Prashant Chaudhary, Pavan Kurariya, Jahnavi Bodhankar et al.· NLP & Big Data· 0 citations
Text which has been translated from another language tends to carry with it evidence of translation$\unicode{x2014}$hence, it is often referred to as $\textit{translationese}$. Multilingual large language models (MLLMs) generate text in a variety of languages. However, it is still unclear if MLLMs'generations resemble internal translation (from English or, potentially, other languages) and, thus, result in translationese. Here, we ask the following research questions: (1) Does text generated by MLLMs resemble translationese? (2) How does translationese produced by MLLMs differ from translationese produced through direct translation? We leverage established indicators of translated text to evaluate text generated by state-of-the-art MLLMs in five languages, comparing to both non-translated and human-written baselines in order to isolate translationese from other kinds of interference. Through the use of high-accuracy classification models, analyses of variance on individual linguistic features, and the collection of human annotations in a subset of two languages (German and Spanish), we assess the translationese content of MLLM generations and examine the key features that distinguish MLLM-generated text from typical translation-related interference.
Maria R. Valentini, Téa Wright, Julisa Granados et al.· 0 citations
This study highlights the importance of integrating in-depth linguistic analysis, contextual semantic modelling, and cultural awareness into natural language processing-based translation systems by combining traditional linguistic insights with computational methods.
Hilal Abdul-Raziq Sadiq, Zaxid Maxmudovich Islamov, R. Matibaeva et al.· Digital Technologies Researc...· 0 citations