mlsanityData to 14 Oct 2026

PAWS-X: A Cross-lingual Adversarial Dataset for Paraphrase Identification

arXiv 1908.11828 ↗First posted 30 Aug 2019cs.CLcited by 264

Not on the map. Why?

Abstract (arXiv)

Most existing work on adversarial data generation focuses on English. For example, PAWS (Paraphrase Adversaries from Word Scrambling) consists of challenging English paraphrase identification pairs from Wikipedia and Quora. We remedy this gap with PAWS-X, a new dataset of 23,659 human translated PAWS evaluation pairs in six typologically distinct languages: French, Spanish, German, Chinese, Japanese, and Korean. We provide baseline numbers for three models with different capacity to capture non-local context and sentence structure, and using different multilingual training and evaluation regimes. Multilingual BERT fine-tuned on PAWS English plus machine-translated data performs the best, with a range of 83.1-90.8 accuracy across the non-English languages and an average accuracy gain of 23% over the next best model. PAWS-X shows the effectiveness of deep, multilingual pre-training while also leaving considerable headroom as a new challenge to drive multilingual research that better captures structure and contextual information.

Key facts. Cited by 264 papers in the map (data through 14 Oct 2026). First posted on arXiv 30 Aug 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 38; 2022: 49; 2023: 51; 2024: 57; 2025: 40; 2026: 29.

264 papers.

Sort:

By area:

  1. 2024Qwen2 Technical Reportcited by 2,120
  2. 2023A Comprehensive Overview of Large Language Modelscited by 648
  3. 2021ByT5: Towards a token-free future with pre-trained byte-to-byte modelscited by 509
  4. 2023MEGA: Multilingual Evaluation of Generative AIcited by 293
  5. 2021Few-shot Learning with Multilingual Language Modelscited by 278
  6. 2021ERNIE 3.0: Large-scale Knowledge Enhanced Pre-training for Language Understanding and Generationcited by 260
  7. 2024Tower: An Open Multilingual Large Language Model for Translation-Related Taskscited by 206
  8. 2023Not All Languages Are Created Equal in LLMs: Improving Multilingual Capability by Cross-Lingual-Thought Promptingcited by 165
  9. 2021Charformer: Fast Character Transformers via Gradient-based Subword Tokenizationcited by 160
  10. 2024Eagle and Finch: RWKV with Matrix-Valued States and Dynamic Recurrencecited by 159
  11. 2022On the Representation Collapse of Sparse Mixture of Expertscited by 157
  12. 2023UniMax: Fairer and more Effective Language Sampling for Large-Scale Multilingual Pretrainingcited by 151
38 more
  1. 2022mGPT: Few-Shot Learners Go Multilingualcited by 142
  2. 2021KLUE: Korean Language Understanding Evaluationcited by 114
  3. 2023Tokenizer Choice For LLM Training: Negligible or Crucial?cited by 108
  4. 2021Measure and Improve Robustness in NLP Models: A Surveycited by 107
  5. 2022Towards Leaving No Indic Language Behind: Building Monolingual Corpora, Benchmark and Models for Indic Languagescited by 102
  6. 2021Reasonable Effectiveness of Random Weighting: A Litmus Test for Multi-Task Learningcited by 101
  7. 2022State-of-the-art generalisation research in NLP: A taxonomy and reviewcited by 98
  8. 2023Cross-lingual Prompting: Improving Zero-shot Chain-of-Thought Reasoning across Languagescited by 97
  9. 2022BLOOM+1: Adding Language Support to BLOOM for Zero-Shot Promptingcited by 90
  10. 2023Do Multilingual Language Models Think Better in English?cited by 88
  11. 2023Spanish Pre-trained BERT Model and Evaluation Datacited by 81
  12. 2021XLM-E: Cross-lingual Language Model Pre-training via ELECTRAcited by 72
  13. 2023BUFFET: Benchmarking Large Language Models for Few-shot Cross-lingual Transfercited by 68
  14. 2024CLIcK: A Benchmark Dataset of Cultural and Linguistic Intelligence in Koreancited by 66
  15. 2023PolyLM: An Open Source Polyglot Large Language Modelcited by 64
  16. 2022Overcoming Catastrophic Forgetting in Zero-Shot Cross-Lingual Generationcited by 62
  17. 2022AlexaTM 20B: Few-Shot Learning Using a Large-Scale Multilingual Seq2Seq Modelcited by 57
  18. 2023MEGAVERSE: Benchmarking Large Language Models Across Languages, Modalities, Models and Taskscited by 57
  19. 2024Multilingual Large Language Model: A Survey of Resources, Taxonomy and Frontierscited by 51
  20. 2024LLMs Beyond English: Scaling the Multilingual Capability of LLMs with Cross-Lingual Feedbackcited by 51
  21. 2021A Primer on Pretrained Multilingual Language Modelscited by 49
  22. 2021ERNIE 3.0 Titan: Exploring Larger-scale Knowledge Enhanced Pre-training for Language Understanding and Generationcited by 46
  23. 2021MURAL: Multimodal, Multitask Retrieval Across Languagescited by 44
  24. 2021MT6: Multilingual Pretrained Text-to-Text Transformer with Translation Pairscited by 41
  25. 2021Improving Pretrained Cross-Lingual Language Models via Self-Labeled Word Alignmentcited by 41
  26. 2021Consistency Regularization for Cross-Lingual Fine-Tuningcited by 41
  27. 2023Revisiting Machine Translation for Cross-lingual Classificationcited by 41
  28. 2021Multi-view Subword Regularizationcited by 40
  29. 2024Breaking the Curse of Multilinguality with Cross-lingual Expert Language Modelscited by 38
  30. 2021MarIA: Spanish Language Modelscited by 36
  31. 2024Mitigating Catastrophic Forgetting in Language Transfer via Model Mergingcited by 36
  32. 2025KaLM-Embedding-V2: Superior Training Techniques and Data Inspire A Versatile Embedding Modelcited by 36
  33. 2024Understanding Cross-Lingual Alignment -- A Surveycited by 34
  34. 2023Improving In-context Learning of Multilingual Generative Language Models with Cross-lingual Alignmentcited by 33
  35. 2023Are Multilingual LLMs Culturally-Diverse Reasoners? An Investigation into Multicultural Proverbs and Sayingscited by 32
  36. 2024Is Translation All You Need? A Study on Solving Multilingual Tasks with Large Language Modelscited by 31
  37. 2024MAGNET: Improving the Multilingual Fairness of Language Models with Adaptive Gradient-Based Tokenizationcited by 31
  38. 2025EuroBERT: Scaling Multilingual Encoders for European Languagescited by 31

50 most cited of 264 All 264 (JSON)