mlsanityData to 7 Oct 2026

Multilingual Machine Translation with Open Large Language Models at Practical Scale: An Empirical Study

arXiv 2502.02481 ↗First posted 4 Feb 2025cs.CLcited by 38

Not on the map. Why?

Abstract (arXiv)

Large language models (LLMs) have shown continuously improving multilingual capabilities, and even small-scale open-source models have demonstrated rapid performance enhancement. In this paper, we systematically explore the abilities of open LLMs with less than ten billion parameters to handle multilingual machine translation (MT) tasks. We conduct comprehensive evaluations on six popular LLMs and find that models like Gemma2-9B exhibit impressive multilingual translation capabilities. We then introduce the Parallel-First Monolingual-Second (PFMS) data mixing strategy in the continual pretraining stage to further enhance the MT performance and present GemmaX2-28, a 9B model achieving top-tier multilingual translation performance across 28 languages. Specifically, GemmaX2-28 consistently outperforms the state-of-the-art (SOTA) models such as TowerInstruct and XALMA and achieves competitive performance with Google Translate and GPT-4-turbo.

Key facts. Cited by 38 papers in the map (data through 7 Oct 2026). Builds on 32 of its 49 references that are in the map. First posted on arXiv 4 Feb 2025.

Builds on

32 of 49 references.

References · 32

  1. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 13,958
  2. 2022OPT: Open Pre-trained Transformer Language Modelscited by 3,464
  3. 2018A Call for Clarity in Reporting BLEU Scorescited by 1,790
  4. 2022No Language Left Behind: Scaling Human-Centered Machine Translationcited by 1,290
  5. 2020COMET: A Neural Framework for MT Evaluationcited by 1,024
  6. 2016Bag of Tricks for Efficient Text Classificationcited by 869
  7. 2020The State and Fate of Linguistic Diversity and Inclusion in the NLP Worldcited by 789
  8. 2021The FLORES-101 Evaluation Benchmark for Low-Resource and Multilingual Machine Translationcited by 556
  9. 2023The Falcon Series of Open Language Modelscited by 526
  10. 2024Contrastive Preference Optimization: Pushing the Boundaries of LLM Performance in Machine Translationcited by 390
  11. 2016FastText.zip: Compressing text classification modelscited by 331
  12. 2023Efficient and Effective Text Encoding for Chinese LLaMA and Alpacacited by 302
20 more
  1. 2024Aya Model: An Instruction Finetuned Open-Access Multilingual Language Modelcited by 292
  2. 2021Few-shot Learning with Multilingual Language Modelscited by 278
  3. 2023xCOMET: Transparent Machine Translation Evaluation through Fine-grained Error Detectioncited by 260
  4. 2024Tower: An Open Multilingual Large Language Model for Translation-Related Taskscited by 205
  5. 2023MADLAD-400: A Multilingual And Document-Level Large Audited Datasetcited by 180
  6. 2019CCMatrix: Mining Billions of High-Quality Parallel Sentences on the WEBcited by 177
  7. 2022In-context Examples Selection for Machine Translationcited by 172
  8. 2023CulturaX: A Cleaned, Enormous, and Multilingual Dataset for Large Language Models in 167 Languagescited by 172
  9. 2019CCAligned: A Massive Collection of Cross-Lingual Web-Document Pairscited by 136
  10. 2024Aya Expanse: Combining Research Breakthroughs for a New Multilingual Frontiercited by 121
  11. 2021Language Models are Few-shot Multilingual Learnerscited by 92
  12. 2023Scaling up COMETKIWI: Unbabel-IST 2023 Submission for the Quality Estimation Shared Taskcited by 88
  13. 2023BUFFET: Benchmarking Large Language Models for Few-shot Cross-lingual Transfercited by 68
  14. 2023Eliciting the Translation Ability of Large Language Models via Multilingual Finetuning with Translation Instructionscited by 55
  15. 2023OpenICL: An Open-Source Framework for In-context Learningcited by 51
  16. 2024LLaMAX: Scaling Linguistic Horizons of LLM by Enhancing Translation Capabilities Beyond 100 Languagescited by 47
  17. 2024X-ALMA: Plug & Play Modules and Adaptive Rejection for Quality Translation at Scalecited by 36
  18. 2023Improving In-context Learning of Multilingual Generative Language Models with Cross-lingual Alignmentcited by 33
  19. 2024Efficiently Exploring Large Language Models for Document-Level Machine Translation with In-context Learningcited by 23
  20. 2023Learning Multilingual Sentence Representations with Cross-lingual Consistency Regularizationcited by 7

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 20; 2026: 18.

38 papers.

Sort:

By area:

  1. 2025Tower+: Bridging Generality and Translation Specialization in Multilingual LLMscited by 40
  2. 2025SSR-Zero: Simple Self-Rewarding Reinforcement Learning for Machine Translationcited by 13
  3. 2025Bridging the Linguistic Divide: A Survey on Leveraging Large Language Models for Machine Translationcited by 8
  4. 2025Hunyuan-MT Technical Reportcited by 7
  5. 2025Advancing Dialectal Arabic to Modern Standard Arabic Machine Translationcited by 6
  6. 2025CCFQA: A Benchmark for Cross-Lingual and Cross-Modal Speech and Text Factuality Evaluationcited by 5
  7. 2026Multilingual Amnesia: On the Transferability of Unlearning in Multilingual LLMscited by 5
  8. 2025KIT's Offline Speech Translation and Instruction Following Submission for IWSLT 2025cited by 4
  9. 2025NiuTrans.LMT: Toward Inclusive and Scalable Multilingual Machine Translation with LLMscited by 4
  10. 2026Scaling Model and Data for Multilingual Machine Translation with Open Large Language Modelscited by 4
  11. 2025M2PO: Multi-Perspective Multi-Pair Preference Optimization for Machine Translationcited by 3
  12. 2025MCAT: Scaling Many-to-Many Speech-to-Text Translation with MLLMs to 70 Languagescited by 3
26 more
  1. 2025Finetune-RAG: Fine-Tuning Language Models to Resist Hallucination in Retrieval-Augmented Generationcited by 2
  2. 2025Mutarjim: Advancing Bidirectional Arabic-English Translation with a Small Language Modelcited by 2
  3. 2025Translation in the Wildcited by 2
  4. 2026XProvence: Zero-Cost Multilingual Context Pruning for Retrieval-Augmented Generationcited by 2
  5. 2026Benchmarking Machine Translation on Chinese Social Media Textscited by 2
  6. 2026JobResQA: A Benchmark for LLM Machine Reading Comprehension on Multilingual R'esum'es and JDscited by 2
  7. 2026RouteLMT: Learned Sample Routing for Hybrid LLM Translation Deploymentcited by 2
  8. 2025VietMix: A Naturally-Occurring Parallel Corpus and Augmentation Framework for Vietnamese-English Code-Mixed Machine Translationcited by 1
  9. 2025Beyond the Cloud: Assessing the Benefits and Drawbacks of Local LLM Deployment for Translatorscited by 1
  10. 2025Evaluating the Impact of Verbal Multiword Expressions on Machine Translationcited by 1
  11. 2025HY-MT1.5 Technical Reportcited by 1
  12. 2026CLewR: Curriculum Learning with Restarts for Machine Translation Preference Learningcited by 1
  13. 2026MERIT: Multilingual Expert-Reward Informed Tuning for Chinese-Centric Low-Resource Machine Translationcited by 1
  14. 2026Mitigating Cross-Lingual Cultural Inconsistencies in LLMs via Consensus-Driven Preference Optimisationcited by 1
  15. 2025QUST_NLP at SemEval-2025 Task 7: A Three-Stage Retrieval Framework for Monolingual and Crosslingual Fact-Checked Claim Retrievalcited by 0
  16. 2025RL from Teacher-Model Refinement: Gradual Imitation Learning for Machine Translationcited by 0
  17. 2025OLaPh: Optimal Language Phonemizercited by 0
  18. 2026DIETA: A Decoder-only transformer-based model for Italian-English machine TrAnslationcited by 0
  19. 2026Scalable Multilingual Multimodal Machine Translation with Speech-Text Fusioncited by 0
  20. 2026Language Bias under Conflicting Information in Multilingual LLMscited by 0
  21. 2026Translation Analytics for Freelancers II: Benchmarking Local LLMs for Confidential Translation Workflowscited by 0
  22. 2026Beyond Semantics: Modeling Factual and Affective Perceptual Experiences from Vision-Language Datacited by 0
  23. 2026Sycophancy as a Multilingual Alignment Failure: How Safety Degrades Across Languages, Topics, and Modelscited by 0
  24. 2026CAT-Translate: Building Compact Open-Source Models for Japanese-English Translationcited by 0
  25. 2026Reference-Free Post-Training of Open Large Language Models for Multilingual Machine Translationcited by 0
  26. 2026Reference-Grounded Data Curation for Instruction-Following Thai-English Machine Translationcited by 0