Jul 2026· International Journal of Advanced Research in Science, Communication and Technology· 0 citations· 6 references
TL;DR
The results indicate that a moderately sized, shared self-attention architecture can deliver production-quality multilin-gual translation within the resource constraints of an academic de-ployment, while surfacing clear directions – low-resource language coverage, domain adaptation, and speech-based extension – for con-tinued development.
Abstract
Cross-lingual communication remains constrained by translation tools that fail to preserve sentence-level meaning, par-ticularly for idiomatic phrasing, long or multi-clause sentences, and low-resource languages. This paper presents the design and evaluation of a browser-accessible multilingual translation plat-form built around a Transformer encoder–decoder network. A sin-gle shared model is fine-tuned across five languages – English, Hindi, French, Spanish, and German – using subword vocabu-lary construction, automatic source-language identification, and an attention-based explainability layer that exposes token-level align-ment through heatmap visualization. The system was evaluated on a held-out multilingual test set using BLEU, ROUGE-L, and accuracy metrics, achieving a macro-averaged translation accuracy of 92.7%, exceeding a 90% target and outperforming LSTM, GRU, statistical, and rule-based baselines by a wide margin. Average inference la-tency remained below 185 milliseconds per request, supporting in-teractive use. The results indicate that a moderately sized, shared self-attention architecture can deliver production-quality multilin-gual translation within the resource constraints of an academic de-ployment, while surfacing clear directions – low-resource language coverage, domain adaptation, and speech-based extension – for con-tinued development
Language barriers continue to limit communication, education, and access to information across global digital platforms. Conventional rule-based and statistical machine translation systems frequently fail to capture sentence-level context, grammar, and semantic meaning, producing inaccurate or unnatural translations, particularly for idiomatic expressions and morphologically complex languages. This paper presents an automated multilingual translation system built around a Transformer-based Neural Machine Translation (NMT) model that leverages multi-head self-attention to translate text between English, Hindi, French, Spanish, and German with contextual awareness. The system integrates automatic source-language detection, a text preprocessing pipeline (cleaning, normalization, subword tokenization), attention-based translation explainability, and a web-based interface supporting real-time translation, translation-history storage, and downloadable reports. The Transformer model is fine-tuned on multilingual parallel corpora drawn from the OPUS and WMT repositories using an 80:10:10 train/validation/test split. Evaluation on a held-out multilingual test set using BLEU, ROUGE-L, and translation-accuracy metrics shows a macro-average translation accuracy of 92.7%, exceeding a 90% target, with a mean inference time of 178.5 ms per request. Comparative evaluation against LSTM-based, GRU-based, statistical, and rule-based baselines shows the proposed Transformer model outperforming all four alternatives on both accuracy and BLEU score. These results indicate that combining attention-based Transformer translation with practical deployment features — language detection, explainability, history management, and reporting — can deliver an accurate, scalable, and user-accessible multilingual translation platform.
Chinmaya Md, Supreetha Hd· International journal of tec...· 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
This work introduces M-GATE (Multilingual Grammar, Accuracy in Translation, and Efficiency), a benchmark of linguistic proficiency spanning 30 typologically diverse languages from high- to low-resource, and evaluates over 50 models in more than 80 configurations.
Tomáš Burkert, Angelika Peljak-Łapińska, David Zelený· 0 citations
A PMI-based translation metric is proposed, which is less dependent on the target language and correlates strongly with chrF, and finds that CLA with English predicts translation quality comparably to or better than source-target CLA.
Adnan Al Ali, Kathy Hämmerl, Jindrich Libovický et al.· 0 citations
The adoption of large pre-trained multilingual models for neural machine translation (MNMT) faces a major challenge: excessive memory and computational consumption due to overly large vocabularies and embedding layers. Although existing compression methods like pruning, quantization and knowledge distillation reduce parameter redundancy, they mainly preserve the structure of the original vocabulary, thereby leaving a major source of inefficiency unresolved. We propose in this paper a general optimization framework that combines a vocabulary pruning method with a targeted fine-tuning protocol for MNMT models. We evaluate the proposed framework using three models (M2M100, NLLB-200, mBART-50) on the English-Arabic language pair. Our approach reduces the vocabulary size from over 128,000 to approximately 10,000 tokens, enabling a 60% memory saving without any loss in performance. Results show that optimized multilingual models can match or exceed the performance of dedicated bilingual baselines. In particular, the pruned and fine-tuned M2M100 model achieves a competitive BLEU score of 42.04 (against 44.59 for the OPUS-MTen- ar bilingual model) while it significantly outperforms it on the COMET metric (0.8730 vs 0.7911) revealing superior semantic adequacy and fluency.
Ahmed Amine Aliane, N. Semmar, H. Aliane· 0 citations
Evaluation of multilingual large language models has grown rapidly in recent years, yet Marathi, spoken by over 83 million people across India, has received almost no systematic probing beyond surface-level benchmark tests. Most existing multilingual evaluations either omit Marathi entirely or rely on machine-translated test sets that fail to capture the morphological complexity that de-fines the language. We evaluate four models, namely Llama-3.1-8B, Llama-3.3-70B, Mistral-7B, and Qwen3-32B, on our manually curated Marathi dataset across three probing dimensions: Devanagari versus Roman-ized script, Marathi-English code-mixing, and syntactic structures including SOV word order, vibhakti case markers, verb gender agreement, and postpositions. Models are tested under English and Marathi instruction conditions across translation, similarity, grammat-icality, and case marker tasks. Translation quality is evaluated using both token-level F1 and BERTScore to capture paraphrase equivalence beyond surface word overlap. All models drop between 7.9% and 20.5% on Roman-ized input. The negative subjunctive marker nasta is ignored by every model. Vibhakti case markers are consistently replaced with Hindi equivalents, revealing that multilingual training has not produced separate internal representations for Hindi and Marathi despite their distinct morphological systems. These findings reveal structural gaps in how current multilingual LLMs handle morphologically rich, low-resource Indic languages and point to spe-cific areas where dedicated Marathi pretraining data would most benefit future work.
Tejas M Patil, Barnali Chetia· Proceedings of the 1st Works...· 0 citations