Evaluating LLMs' Multilingual Capabilities for Bengali: Benchmark Creation and Performance Analysis

TL;DR

Constructed Bengali multilingual benchmarks, evaluated 10 open-source LLMs, found size and tokenization impact performance, with larger models showing more robustness.

cs.CL πŸ”΄ Advanced 2025-07-31 11 citations 45 views
Shimanto Bhowmik Tawsif Tashwar Dipto Md Sazzad Islam Sheryl Hsu Tahsin Reasat
multilingual NLP benchmark creation model evaluation tokenization low-resource languages

Key Findings

Methodology

The study employed systematic translation of 8 high-quality English datasets into Bengali using GPT-4 mini, followed by manual validation to ensure translation fidelity. Based on these datasets, ten open-source multilingual LLMs were evaluated across various tasks, analyzing their error modes and tokenization behaviors. Metrics such as accuracy, response error rate, and model consistency were used to compare performance in English and Bengali, with particular focus on model size and architecture effects. The evaluation framework combined automated metrics and human-like judgment via GPT-based evaluators, providing comprehensive insights into language-specific performance gaps.

Key Results

  • Models exhibited significantly lower average accuracy in Bengali compared to English, especially smaller models and Mistral family, with the lowest accuracy at 0.048 (Mistral 7B) and highest at 0.966 (Qwen 72B). Larger models demonstrated more stable performance, with DeepSeek architecture showing notable robustness. Some models even outperformed English prompts in structured tasks, indicating the influence of translation quality and tokenization strategies.
  • Tokenization analysis revealed that Bengali inputs produce approximately 30% more tokens per word than English, with token efficiency inversely correlated with model performance. Complex multi-character glyphs in Bengali cause token fragmentation, impairing model understanding and reasoning. Efficient tokenization correlates with higher accuracy, emphasizing the importance of optimized subword segmentation.
  • High-quality translation using GPT-4 mini was crucial for reliable evaluation, as it preserved semantic integrity better than other machine translation tools. This approach demonstrates a practical pathway for creating benchmarks in low-resource languages, highlighting the critical role of translation fidelity in model assessment. The publicly released datasets and tools facilitate future research in underrepresented languages.

Significance

This work addresses the critical gap in standardized benchmarks for Bengali NLP, providing a comprehensive evaluation framework that enables fair comparison of models across languages. By analyzing the impact of model size and tokenization, it offers empirical evidence guiding model design for low-resource languages. The findings promote equitable access to advanced NLP technologies, fostering inclusive language AI development. The methodology and datasets serve as a foundation for future multilingual research, encouraging community efforts to improve low-resource language NLP capabilities and ensuring broader linguistic diversity in AI applications.

Technical Contribution

The study introduced a systematic pipeline combining GPT-4 mini translation and manual validation to produce high-quality Bengali datasets. It evaluated ten open-source multilingual LLMs across multiple tasks, integrating detailed tokenization analysis to reveal the relationship between token efficiency and model performance. The research also provided a comparative analysis of model robustness across languages, highlighting architecture-specific strengths like DeepSeek. These contributions advance the state-of-the-art in low-resource multilingual evaluation, emphasizing data quality and tokenization strategies as key factors for performance improvements.

Novelty

This is the first comprehensive effort to systematically translate and evaluate multiple NLP benchmarks in Bengali, utilizing GPT-4 mini for high-quality translation and combining automated and manual validation. Unlike prior works that rely solely on machine translation or limited datasets, this study emphasizes data quality, detailed tokenization analysis, and cross-lingual performance comparison. Its innovative approach provides a holistic view of low-resource language model capabilities, setting a new standard for multilingual benchmark development and evaluation.

Limitations

  • The translation process depends heavily on GPT-4 mini, which may introduce semantic biases or cultural inaccuracies, potentially affecting the representativeness of the datasets.
  • Evaluation is limited to open-source models, excluding commercial or proprietary models, which may perform differently in real-world applications.
  • Tokenization analysis focuses on static metrics and does not fully capture dynamic behaviors during inference, suggesting the need for more integrated evaluation methods.

Future Work

Future research will expand to include more low-resource languages, develop adaptive tokenization techniques tailored to complex scripts, and incorporate human-in-the-loop evaluation for better semantic fidelity. Exploring multi-modal data integration and fine-tuning strategies will further enhance model robustness. Additionally, establishing community-driven benchmarks and expanding dataset diversity will promote inclusive NLP development for underrepresented languages.

AI Executive Summary

This research presents a pioneering effort to establish standardized multilingual benchmarks for Bengali, a language historically underrepresented in NLP. By translating eight high-quality English datasets into Bengali using GPT-4 mini and validating the translations through manual review, the study ensures data reliability. The evaluation of ten open-source multilingual large language models across diverse tasks reveals a consistent performance gap between Bengali and English, with smaller models and certain architectures like Mistral showing significant weaknesses. Notably, larger models such as Qwen 72B and DeepSeek demonstrate more stable performance, highlighting the importance of model scale and architecture in cross-lingual robustness.

A key insight from the analysis is the inverse relationship between tokenization efficiency and model accuracy. Bengali inputs tend to generate more tokens per word, especially with complex glyphs, which hampers model understanding and reasoning. Efficient tokenization strategies, therefore, are critical for improving low-resource language performance. The study also underscores the importance of translation quality; GPT-4 mini's high-fidelity translations enable more accurate benchmarking, setting a new standard for low-resource language evaluation.

Overall, this work significantly advances the field by providing high-quality datasets, comprehensive evaluation metrics, and detailed analysis of model behaviors in Bengali. It offers practical guidance for developing more inclusive NLP models that can serve diverse linguistic communities. Future directions include expanding benchmarks to other low-resource languages, refining tokenization methods, and integrating multi-modal data to further boost model robustness and applicability. This research lays a solid foundation for democratizing NLP technology across languages, fostering a more equitable AI landscape.

Deep Dive

Abstract

Bengali is an underrepresented language in NLP research. However, it remains a challenge due to its unique linguistic structure and computational constraints. In this work, we systematically investigate the challenges that hinder Bengali NLP performance by focusing on the absence of standardized evaluation benchmarks. We then evaluated 10 recent open source Large Language Models (LLMs) in 8 of the translated datasets and performed a comprehensive error analysis to pinpoint their primary failure modes. Our findings reveal consistent performance gaps for Bengali compared to English, particularly for smaller models and specific model families like Mistral. We also identified promising robustness in certain architectures, such as DeepSeek, that maintain more stable performance across languages. Our analysis reveals an inverse relationship between tokenization efficiency and LLM accuracy where models tend to perform worse when inputs are excessively tokenized, whereas more efficient \& concise tokenization results in improved performance. These findings highlight critical areas where current models fall short and underscore the need for improved dataset quality and evaluation methodologies tailored to multilingual contexts. This work will catalyze further research on NLP for underrepresented languages, helping to democratize access to advanced language technologies worldwide. The code and dataset used in this research is publicly available at https://github.com/BengaliAI/bn-llm-benchmark.

cs.CL cs.LG

References (20)

BanglaByT5: Byte-Level Modelling for Bangla

Pramit Bhattacharyya, Arnab Bhattacharya

2025 7 citations ⭐ Influential View Analysis β†’

BEnQA: A Question Answering Benchmark for Bengali and English

Sheikh Shafayat, H. Hasan, Minhajur Rahman Chowdhury Mahim et al.

2024 18 citations ⭐ Influential

A Large Multi-target Dataset of Common Bengali Handwritten Graphemes

Samiul Alam, Tahsin Reasat, Asif Sushmit et al.

2020 26 citations ⭐ Influential View Analysis β†’

How Good is Your Tokenizer? On the Monolingual Performance of Multilingual Language Models

Phillip Rust, Jonas Pfeiffer, Ivan Vulic et al.

2020 452 citations ⭐ Influential View Analysis β†’

Language Model Tokenizers Introduce Unfairness Between Languages

Aleksandar Petrov, Emanuele La Malfa, Philip H. S. Torr et al.

2023 283 citations ⭐ Influential View Analysis β†’

Quantifying Language Models' Sensitivity to Spurious Features in Prompt Design or: How I learned to start worrying about prompt formatting

Melanie Sclar, Yejin Choi, Yulia Tsvetkov et al.

2023 991 citations ⭐ Influential View Analysis β†’

Do All Languages Cost the Same? Tokenization in the Era of Commercial Language Models

Orevaoghene Ahia, Sachin Kumar, Hila Gonen et al.

2023 243 citations ⭐ Influential View Analysis β†’

Judging LLM-as-a-judge with MT-Bench and Chatbot Arena

Lianmin Zheng, Wei-Lin Chiang, Ying Sheng et al.

2023 11296 citations ⭐ Influential View Analysis β†’

Large Language Models Are Not Robust Multiple Choice Selectors

Chujie Zheng, Hao Zhou, Fandong Meng et al.

2023 543 citations ⭐ Influential View Analysis β†’

Association for Computational Linguistics

G. Hirst

2001 173 citations

Rethinking Tokenization for Rich Morphology: The Dominance of Unigram over BPE and Morphological Alignment

Saketh Reddy Vemula, Sandipan Dandapat, D. Sharma et al.

2025 11 citations View Analysis β†’

BanglaBERT: Language Model Pretraining and Benchmarks for Low-Resource Language Understanding Evaluation in Bangla

Abhik Bhattacharjee, Tahmid Hasan, Kazi Samin Mubasshir et al.

2021 347 citations View Analysis β†’

BanglaGPT: A Generative Pretrained Transformer-Based Model for Bangla Language

Md. Shahidul Salim, Hasan Murad, Dola Das et al.

2023 9 citations

BenLLM-Eval: A Comprehensive Evaluation into the Potentials and Pitfalls of Large Language Models on Bengali NLP

M. Kabir, Mohammed Saidul Islam, Md Tahmid Rahman Laskar et al.

2023 38 citations View Analysis β†’

IndicLLMSuite: A Blueprint for Creating Pre-training and Fine-Tuning Datasets for Indian Languages

Mohammed Safi Ur Rahman Khan, Priyam Mehta, A. Sankar et al.

2024 71 citations View Analysis β†’

Khayyam Challenge (PersianMMLU): Is Your LLM Truly Wise to The Persian Language?

Omid Ghahroodi, Marzia Nouri, M. V. Sanian et al.

2024 31 citations View Analysis β†’

Bridging the Bosphorus: Advancing Turkish Large Language Models through Strategies for Low-Resource Language Adaptation and Benchmarking

Emre Can Acikgoz, Mete Erdogan, Deniz Yuret

2024 25 citations View Analysis β†’

xFinder: Large Language Models as Automated Evaluators for Reliable Evaluation

Qingchen Yu, Zifan Zheng, Shichao Song et al.

2024 21 citations View Analysis β†’

Lexically Grounded Subword Segmentation

Jindvrich Libovick'y, JindΕ™ich Helcl

2024 14 citations View Analysis β†’

Testing prompt engineering methods for knowledge extraction from text

Fina Polat, Ilaria Tiddi, Paul Groth

2024 38 citations

Cited By (11)

Quantization Effects on Bangla Language Understanding in Large Language Models: A Systematic Evaluation

2026 ⭐ Influential View Analysis β†’

Creating Multilingual Mental Health Dialogue Datasets: Limits of Persona-Based Localization via Nationality and Language

2026 ⭐ Influential View Analysis β†’

Recent Advances in Bangla NLP: A Survey of Few-Shot, Zero-Shot, and Transfer Learning Techniques

2026 ⭐ Influential

Structural Silence: When AI Infrastructure Fails Speakers of Underrepresented Languages

2026 1 citations View Analysis β†’

Do Small Models Use the Law You Give Them? Context-Injected Fine-Tuning for Legal QA in Bangladesh

BenHalluEval: A Multi-Task Hallucination Evaluation Framework for Large Language Models on Bengali

2026 1 citations View Analysis β†’

Cross-Lingual Sentiment Misalignment: Auditing Multilingual Language Models for Inversion Risk, Dialectal Representation, and Affective Stability

2026 5 citations View Analysis β†’

UrduBench: An Urdu Reasoning Benchmark using Contextually Ensembled Translations with Human-in-the-Loop

2026 2 citations View Analysis β†’

GanitLLM: Difficulty-Aware Bengali Mathematical Reasoning through Curriculum-GRPO

2026 8 citations View Analysis β†’

PESCO-BERT: An Efficient Prompt-Based Contrastive Learning for Bangla News Classification

2025

CUET_Expelliarmus at BLP2025 Task 2: Leveraging Instruction Translation and Refinement for Bangla-to-Python Code Generation with Open-Source LLMs

2025 1 citations