A Comparative Analysis of Retrieval-Augmented Generation Techniques for Bengali Standard-to-Dialect Machine Translation Using LLMs

• URL: http://arxiv.org/abs/2512.14179v1
• Date: Tue, 16 Dec 2025 08:18:18 GMT
• Title: A Comparative Analysis of Retrieval-Augmented Generation Techniques for Bengali Standard-to-Dialect Machine Translation Using LLMs
• Abstract summary: Translating from a standard language to its regional dialects is a significant NLP challenge due to scarce data and linguistic variation.<n>This paper proposes and compares two novel RAG pipelines for standard-to-dialectal Bengali translation.
• Score: 0.3227658251731014
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Translating from a standard language to its regional dialects is a significant NLP challenge due to scarce data and linguistic variation, a problem prominent in the Bengali language. This paper proposes and compares two novel RAG pipelines for standard-to-dialectal Bengali translation. The first, a Transcript-Based Pipeline, uses large dialect sentence contexts from audio transcripts. The second, a more effective Standardized Sentence-Pairs Pipeline, utilizes structured local\_dialect:standard\_bengali sentence pairs. We evaluated both pipelines across six Bengali dialects and multiple LLMs using BLEU, ChrF, WER, and BERTScore. Our findings show that the sentence-pair pipeline consistently outperforms the transcript-based one, reducing Word Error Rate (WER) from 76\% to 55\% for the Chittagong dialect. Critically, this RAG approach enables smaller models (e.g., Llama-3.1-8B) to outperform much larger models (e.g., GPT-OSS-120B), demonstrating that a well-designed retrieval strategy can be more crucial than model size. This work contributes an effective, fine-tuning-free solution for low-resource dialect translation, offering a practical blueprint for preserving linguistic diversity.

Related papers

• Cross Lingual Transfer in Tulu Legal Comprehension: Script-Dependent Improvement and RAG-Induced Knowledge Conflict [0.0]
  Three models were tested on the ability to classify legal complaints written in a low resource Dravidian language (Tulu)<n>Retrieving from a corpus of Kannada legal papers across a RAG framework caused mixed results.<n>Results identify the model's parsing of information and subsequent reasoning as the source of reasoning failure.
  arXiv  Detail & Related papers  (2026-08-13T17:55:45Z)
• Efficient Adaptation of LLMs for Hate Speech Detection in Low-Resource Languages: A Comparative Study on Roman Urdu [0.3436215734071903]
  It is challenging to detect hate speech in Low Resource Languages (LRLs) because of the absence of annotated data, the informality of its language structure, and the lack of standardized grammar.<n>The objective of this paper is to conduct a comprehensive assessment of Large Language Models (LLMs) for Hate Speech Detection (HSD) in Roman Urdu script.<n>Experiments are conducted on the PURUTT (Parallel Urdu and Roman Urdu Corpus for Toxic Comments and Transliteration) dataset with over 72,000 annotated comments.
  arXiv  Detail & Related papers  (2026-08-06T11:28:19Z)
• Benchmarking Bengali Dialectal Bias: A Multi-Stage Framework Integrating RAG-Based Translation and Human-Augmented RLAIF [0.3227658251731014]
  Large language models (LLMs) frequently exhibit performance biases against regional dialects of low-resource languages.<n>We propose a two-phase framework to evaluate dialectal bias in LLM question-answering across nine Bengali dialects.
  arXiv  Detail & Related papers  (2026-03-22T18:44:57Z)
• Raising Bars, Not Parameters: LilMoo Compact Language Model for Hindi [9.65814816271915]
  LilMoo is a 0.6-billion- parameter Hindi language model trained entirely from scratch.<n>It is developed through a fully transparent and reproducible pipeline optimized for limited compute environments.<n>Across comprehensive evaluation suites, LilMoo consistently outperforms comparably sized multilingual baselines.
  arXiv  Detail & Related papers  (2026-03-03T20:31:25Z)
• POTSA: A Cross-Lingual Speech Alignment Framework for Low Resource Speech-to-Text Translation [47.51298472124902]
  We propose a new framework based on cross-lingual parallel speech pairs and Optimal Transport (OT) to bridge high- and low-resource translation gaps.<n>Our method achieves SOTA performance, with +0.93 BLEU on average over five common languages and +5.05 BLEU on zero-shot languages.
  arXiv  Detail & Related papers  (2025-11-12T11:47:56Z)
• A comparison of pipelines for the translation of a low resource language based on transformers [0.0]
  This work compares three pipelines to produce machine translators for Bambara, a Mande language spoken in Africa by about 14,188,850 people.<n>The first pipeline trains a simple transformer to translate sentences from French into Bambara.<n>The second fine-tunes LLaMA3 (3B-8B) instructor models using decoder-only architectures for French-to-Bambara translation.<n>The third pipeline uses language distillation with a student-teacher dual neural network to integrate Bambara into a pre-trained LaBSE model, which provides language-agnostic embeddings.
  arXiv  Detail & Related papers  (2025-09-15T23:36:49Z)
• Vuyko Mistral: Adapting LLMs for Low-Resource Dialectal Translation [0.0]
  This paper introduces the first effort to adapt large language models to the Ukrainian dialect Hutsul.<n>We created a parallel corpus of 9852 dialect-to-standard Ukrainian sentence pairs and a dictionary of 7320 dialectal word mappings.
  arXiv  Detail & Related papers  (2025-06-09T10:30:35Z)
• LLM-Based Evaluation of Low-Resource Machine Translation: A Reference-less Dialect Guided Approach with a Refined Sylheti-English Benchmark [1.3927943269211591]
  We propose a comprehensive framework that enhances Large Language Models (LLMs)-based machine translation evaluation.<n>We extend the ONUBAD dataset by incorporating Sylheti-English sentence pairs, corresponding machine translations, and Direct Assessment (DA) scores annotated by native speakers.<n>Our evaluation shows that the proposed pipeline consistently outperforms existing methods, achieving the highest gain of +0.1083 in Spearman correlation.
  arXiv  Detail & Related papers  (2025-05-18T07:24:13Z)
• Breaking the Script Barrier in Multilingual Pre-Trained Language Models with Transliteration-Based Post-Training Alignment [50.27950279695363]
  The transfer performance is often hindered when a low-resource target language is written in a different script than the high-resource source language. Inspired by recent work that uses transliteration to address this problem, our paper proposes a transliteration-based post-pretraining alignment (PPA) method.
  arXiv  Detail & Related papers  (2024-06-28T08:59:24Z)
• SumTra: A Differentiable Pipeline for Few-Shot Cross-Lingual Summarization [8.971234046933349]
  Cross-lingual summarization (XLS) generates summaries in a language different from that of the input documents. We propose revisiting the summarize-and-translate pipeline, where the summarization and translation tasks are performed in a sequence. This approach allows reusing the many, publicly-available resources for monolingual summarization and translation, obtaining a very competitive zero-shot performance.
  arXiv  Detail & Related papers  (2024-03-20T02:04:42Z)
• Modelling Latent Translations for Cross-Lingual Transfer [47.61502999819699]
  We propose a new technique that integrates both steps of the traditional pipeline (translation and classification) into a single model. We evaluate our novel latent translation-based model on a series of multilingual NLU tasks. We report gains for both zero-shot and few-shot learning setups, up to 2.7 accuracy points on average.
  arXiv  Detail & Related papers  (2021-07-23T17:11:27Z)
• Self-Training Sampling with Monolingual Data Uncertainty for Neural Machine Translation [98.83925811122795]
  We propose to improve the sampling procedure by selecting the most informative monolingual sentences to complement the parallel data. We compute the uncertainty of monolingual sentences using the bilingual dictionary extracted from the parallel data. Experimental results on large-scale WMT English$Rightarrow$German and English$Rightarrow$Chinese datasets demonstrate the effectiveness of the proposed approach.
  arXiv  Detail & Related papers  (2021-06-02T05:01:36Z)
• Rejuvenating Low-Frequency Words: Making the Most of Parallel Data in Non-Autoregressive Translation [98.11249019844281]
  Knowledge distillation (KD) is commonly used to construct synthetic data for training non-autoregressive translation (NAT) models. We propose reverse KD to rejuvenate more alignments for low-frequency target words. Results demonstrate that the proposed approach can significantly and universally improve translation quality.
  arXiv  Detail & Related papers  (2021-06-02T02:41:40Z)
• Continual Mixed-Language Pre-Training for Extremely Low-Resource Neural Machine Translation [53.22775597051498]
  We present a continual pre-training framework on mBART to effectively adapt it to unseen languages. Results show that our method can consistently improve the fine-tuning performance upon the mBART baseline. Our approach also boosts the performance on translation pairs where both languages are seen in the original mBART's pre-training.
  arXiv  Detail & Related papers  (2021-05-09T14:49:07Z)
• Cross-lingual Machine Reading Comprehension with Language Branch Knowledge Distillation [105.41167108465085]
  Cross-lingual Machine Reading (CLMRC) remains a challenging problem due to the lack of large-scale datasets in low-source languages. We propose a novel augmentation approach named Language Branch Machine Reading (LBMRC) LBMRC trains multiple machine reading comprehension (MRC) models proficient in individual language. We devise a multilingual distillation approach to amalgamate knowledge from multiple language branch models to a single model for all target languages.
  arXiv  Detail & Related papers  (2020-10-27T13:12:17Z)
• Investigating Language Impact in Bilingual Approaches for Computational Language Documentation [28.838960956506018]
  This paper investigates how the choice of translation language affects the posterior documentation work. We create 56 bilingual pairs that we apply to the task of low-resource unsupervised word segmentation and alignment. Our results suggest that incorporating clues into the neural models' input representation increases their translation and alignment quality.
  arXiv  Detail & Related papers  (2020-03-30T10:30:34Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.