mlsanityData to 14 Oct 2026

Do Multilingual Language Models Think Better in English?

arXiv 2308.01223 ↗First posted 2 Aug 2023cs.CL cs.AI cs.LGcited by 88

On the map: LLM reasoning → Reasoning benchmarks and evaluation

Abstract (arXiv)

Translate-test is a popular technique to improve the performance of multilingual language models. This approach works by translating the input into English using an external machine translation system, and running inference over the translated input. However, these improvements can be attributed to the use of a separate translation system, which is typically trained on large amounts of parallel data not seen by the language model. In this work, we introduce a new approach called self-translate, which overcomes the need of an external translation system by leveraging the few-shot translation capabilities of multilingual language models. Experiments over 5 tasks show that self-translate consistently outperforms direct inference, demonstrating that language models are unable to leverage their full multilingual potential when prompted in non-English languages. Our code is available at https://github.com/juletx/self-translate.

Key facts. Cited by 88 papers in the map (data through 14 Oct 2026). Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning. Builds on 24 of its 38 references that are in the map. First posted on arXiv 2 Aug 2023.

Builds on

24 of 38 references.

Closest ancestors · 4

Same direction, newest first

  1. 2023PaLM 2 Technical Report17 May
  2. 2023Not All Languages Are Created Equal in LLMs: Improving Multilingual Capability by Cross-Lingual-Thought Prompting11 May
  3. 2023LLaMA: Open and Efficient Foundation Language Models27 Feb
  4. 2023A Multitask, Multilingual, Multimodal Evaluation of ChatGPT on Reasoning, Hallucination, and Interactivity8 Feb

Other references · 20

  1. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 13,028
  2. 2022PaLM: Scaling Language Modeling with Pathwayscited by 5,326
  3. 2022Emergent Abilities of Large Language Modelscited by 2,512
  4. 2022BLOOM: A 176B-Parameter Open-Access Multilingual Language Modelcited by 1,984
  5. 2022No Language Left Behind: Scaling Human-Centered Machine Translationcited by 1,312
  6. 2018XNLI: Evaluating Cross-lingual Sentence Representationscited by 915
  7. 2022Language Models are Multilingual Chain-of-Thought Reasonerscited by 530
  8. 2023How Good Are GPT Models at Machine Translation? A Comprehensive Evaluationcited by 348
  9. 2020XCOPA: A Multilingual Dataset for Causal Commonsense Reasoningcited by 333
  10. 2023MEGA: Multilingual Evaluation of Generative AIcited by 293
  11. 2021Few-shot Learning with Multilingual Language Modelscited by 278
  12. 2019PAWS-X: A Cross-lingual Adversarial Dataset for Paraphrase Identificationcited by 264
8 more
  1. 2023Prompting Large Language Model for Machine Translation: A Case Studycited by 250
  2. 2022Prompting PaLM for Translation: Assessing Strategies and Performancecited by 145
  3. 2023Don't Trust ChatGPT when Your Question is not in English: A Study of Multilingual Abilities and Types of LLMscited by 104
  4. 2020Translation Artifacts in Cross-lingual Transfer Learningcited by 96
  5. 2023PolyLM: An Open Source Polyglot Large Language Modelcited by 64
  6. 2023Revisiting Machine Translation for Cross-lingual Classificationcited by 41
  7. 2021Modelling Latent Translations for Cross-Lingual Transfercited by 16
  8. 2022On the Role of Parallel Data in Cross-lingual Transfer Learningcited by 11

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 31; 2025: 34; 2026: 23.

88 papers.

Sort:

By area:

  1. 2024Do Llamas Work in English? On the Latent Language of Multilingual Transformerscited by 281
  2. 2025MMLU-ProX: A Multilingual Benchmark for Advanced Large Language Model Evaluationcited by 79
  3. 2025Do Multilingual LLMs Think In English?cited by 78
  4. 2024Multilingual Large Language Model: A Survey of Resources, Taxonomy and Frontierscited by 51
  5. 2024LLMs Beyond English: Scaling the Multilingual Capability of LLMs with Cross-Lingual Feedbackcited by 51
  6. 2024Enhancing Multilingual Capabilities of Large Language Models through Self-Distillation from Resource-Rich Languagescited by 36
  7. 2024Language Imbalance Driven Rewarding for Multilingual Self-improvingcited by 28
  8. 2024INDIC QA BENCHMARK: A Multilingual Benchmark to Evaluate Question Answering capability of LLMs for Indic Languagescited by 24
  9. 2024BertaQA: How Much Do Language Models Know About Local Culture?cited by 23
  10. 2025The Emergence of Abstract Thought in Large Language Models Beyond Any Languagecited by 21
  11. 2024Fine-tuning Large Language Models with Sequential Instructionscited by 20
  12. 2025MPO: Multilingual Safety Alignment via Reward Gap Optimizationcited by 18
38 more
  1. 2024Evaluating the Capabilities of Large Language Models for Multi-label Emotion Understandingcited by 17
  2. 2024No Filter: Cultural and Socioeconomic Diversity in Contrastive Vision-Language Modelscited by 16
  3. 2025Multi-lingual Multi-turn Automated Red Teaming for LLMscited by 13
  4. 2024Breaking the Language Barrier: Can Direct Inference Outperform Pre-Translation in Multilingual LLM Applications?cited by 12
  5. 2025Can you map it to English? The Role of Cross-Lingual Alignment in Multilingual Performance of LLMscited by 12
  6. 2024Zero-shot cross-lingual transfer in instruction tuning of large language modelscited by 11
  7. 2024AL-QASIDA: Analyzing LLM Quality and Accuracy Systematically in Dialectal Arabiccited by 11
  8. 2025Improving Multilingual Language Models by Aligning Representations through Steeringcited by 11
  9. 2025Do LLMs Understand Your Translations? Evaluating Paragraph-level MT with Question Answeringcited by 10
  10. 2025X-WebAgentBench: A Multilingual Interactive Web Benchmark for Evaluating Global Agentic Systemcited by 10
  11. 2025The Translation Barrier Hypothesis: Multilingual Generation with Large Language Models Suffers from Implicit Translation Failurecited by 10
  12. 2025Mind the Gap... or Not? How Translation Errors and Evaluation Details Skew Multilingual Resultscited by 10
  13. 20241+1>2: Can Large Language Models Serve as Cross-Lingual Knowledge Aggregators?cited by 9
  14. 2024Exploring the Role of Transliteration in In-Context Learning for Low-resource Languages Written in Non-Latin Scriptscited by 8
  15. 2025Just Go Parallel: Improving the Multilingual Capabilities of Large Language Modelscited by 8
  16. 2025Tracing Multilingual Representations in LLMs with Cross-Layer Transcoderscited by 8
  17. 2024English Prompts are Better for NLI-based Zero-Shot Emotion Classification than Target-Language Promptscited by 7
  18. 2024Polarization of Autonomous Generative AI Agents Under Echo Chamberscited by 7
  19. 2024The Roles of English in Evaluating Multilingual Language Modelscited by 7
  20. 2025MAPS: A Multilingual Benchmark for Agent Performance and Securitycited by 7
  21. 2024Lens: Rethinking Multilingual Enhancement for Large Language Modelscited by 6
  22. 2025Exploring Robustness of Multilingual LLMs on Real-World Noisy Datacited by 6
  23. 2025Multilingual Prompt Engineering in Large Language Models: A Survey Across NLP Taskscited by 6
  24. 2025Relic: Enhancing Reward Model Generalization for Low-Resource Indic Languages with Few-Shot Examplescited by 6
  25. 2025Do Political Opinions Transfer Between Western Languages? An Analysis of Unaligned and Aligned Multilingual LLMscited by 6
  26. 2026Gained in Translation: Privileged Pairwise Judges Enhance Multilingual Reasoningcited by 6
  27. 2026Multilingual Safety Alignment Via Sparse Weight Editingcited by 6
  28. 2024Multilingual Pretraining Using a Large Corpus Machine-Translated from a Single Source Languagecited by 5
  29. 2024PromptRefine: Enhancing Few-Shot Performance on Low-Resource Indic Languages with Example Selection from Related Example Bankscited by 5
  30. 2025Document Intelligence in the Era of Large Language Models: A Surveycited by 5
  31. 2026Cross-Lingual Activation Steering for Multilingual Language Modelscited by 5
  32. 2025CoCo-CoLa: Evaluating and Improving Language Adherence in Multilingual LLMscited by 4
  33. 2025Understanding LLMs' Cross-Lingual Context Retrieval: How Good It Is And Where It Comes Fromcited by 4
  34. 2025Language Specific Knowledge: Do Models Know Better in X than in English?cited by 4
  35. 2025Read it in Two Steps: Translating Extremely Low-Resource Languages with Code-Augmented Grammar Bookscited by 4
  36. 2025The Cultural Gene of Large Language Models: A Study on the Impact of Cross-Corpus Training on Model Values and Biasescited by 4
  37. 2025Translate, then Detect: Leveraging Machine Translation for Cross-Lingual Toxicity Classificationcited by 4
  38. 2025Beg to Differ: Understanding Reasoning-Answer Misalignment Across Languagescited by 4

50 most cited of 88 All 88 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2023The unreasonable effectiveness of few-shot learning for machine translation
  2. 2026Recovered in Translation: Efficient Pipeline for Automated Translation of Benchmarks and Datasets
  3. 2025Trans-Zero: Self-Play Incentivizes Large Language Models for Multilingual Translation Without Parallel Data
  4. 2026The Interlingua Hypothesis: LLMs Translate via a Latent Task-agnostic Feature Space
  5. 2024Getting More from Less: Large Language Models are Good Spontaneous Multilingual Learners
  6. 2023Language Versatilists vs. Specialists: An Empirical Revisiting on Multilingual Transfer Ability
  7. 2024GenTranslate: Large Language Models are Generative Multilingual Speech and Machine Translators
  8. 2024What do Large Language Models Need for Machine Translation Evaluation?