mlsanityData to 7 Oct 2026

Multilingual Pixel Representations for Translation and Effective Cross-lingual Transfer

arXiv 2305.14280 ↗First posted 23 May 2023cs.CLcited by 22

Not on the map. Why?

Abstract (arXiv)

We introduce and demonstrate how to effectively train multilingual machine translation models with pixel representations. We experiment with two different data settings with a variety of language and script coverage, demonstrating improved performance compared to subword embeddings. We explore various properties of pixel representations such as parameter sharing within and across scripts to better understand where they lead to positive transfer. We observe that these properties not only enable seamless cross-lingual transfer to unseen scripts, but make pixel representations more data-efficient than alternatives such as vocabulary expansion. We hope this work contributes to more extensible multilingual models for all languages and scripts.

Key facts. Cited by 22 papers in the map (data through 7 Oct 2026). Builds on 30 of its 45 references that are in the map. First posted on arXiv 23 May 2023.

Builds on

30 of 45 references.

References · 30

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,339
  3. 2019Parameter-Efficient Transfer Learning for NLPcited by 4,195
  4. 2019Unsupervised Cross-lingual Representation Learning at Scalecited by 3,907
  5. 2018SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processingcited by 1,973
  6. 2022No Language Left Behind: Scaling Human-Centered Machine Translationcited by 1,290
  7. 2018Subword Regularization: Improving Neural Network Translation Models with Multiple Subword Candidatescited by 632
  8. 2020Beyond English-Centric Multilingual Machine Translationcited by 559
  9. 2021ByT5: Towards a token-free future with pre-trained byte-to-byte modelscited by 504
  10. 2020MAD-X: An Adapter-Based Framework for Multi-Task Cross-Lingual Transfercited by 443
  11. 2019Massively Multilingual Neural Machine Translation in the Wild: Findings and Challengescited by 253
  12. 2019Representation Degeneration Problem in Training Natural Language Generation Modelscited by 250
18 more
  1. 2019Massively Multilingual Neural Machine Translationcited by 246
  2. 2021CANINE: Pre-training an Efficient Tokenization-Free Encoder for Language Representationcited by 229
  3. 2021Between words and characters: A Brief History of Open-Vocabulary Modeling and Tokenization in NLPcited by 130
  4. 2018When and Why are Pre-trained Word Embeddings Useful for Neural Machine Translation?cited by 125
  5. 2019Neural Machine Translation with Byte-Level Subwordscited by 115
  6. 2020Deep Encoder, Shallow Decoder: Reevaluating Non-autoregressive Machine Translationcited by 104
  7. 2023XLM-V: Overcoming the Vocabulary Bottleneck in Multilingual Masked Language Modelscited by 88
  8. 2020Improving Multilingual Models with Language-Clustered Vocabulariescited by 53
  9. 2021Robust Open-Vocabulary Translation from Visual Text Representationscited by 47
  10. 2016Very Deep Convolutional Networks for End-to-End Speech Recognitioncited by 34
  11. 2018Revisiting Character-Based Neural Machine Translation with Capacity and Compressioncited by 28
  12. 2019Improving Pre-Trained Multilingual Models with Vocabulary Expansioncited by 27
  13. 2015Multilingual Language Processing From Bytescited by 26
  14. 2022Causes and Cures for Interference in Multilingual Translationcited by 23
  15. 2021Alternative Input Signals Ease Transfer in Multilingual Machine Translationcited by 14
  16. 2022Multilingual Neural Machine Translation with Deep Encoder and Multiple Shallow Decoderscited by 12
  17. 2019A Universal Parent Model for Low-Resource Neural Machine Translation Transfercited by 10
  18. 2021Efficient Inference for Multilingual Neural Machine Translationcited by 10

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 3; 2024: 10; 2025: 6; 2026: 3.

22 papers.

Sort:

By area:

  1. 2024MAGNET: Improving the Multilingual Fairness of Language Models with Adaptive Gradient-Based Tokenizationcited by 31
  2. 2023LLM Performance Predictors are good initializers for Architecture Searchcited by 29
  3. 2024Evaluating the Elementary Multilingual Capabilities of Large Language Models with MultiQcited by 27
  4. 2023Text Rendering Strategies for Pixel Language Modelscited by 22
  5. 2024Efficiently Democratizing Medical LLMs for 50 Languages via a Mixture of Language Family Expertscited by 17
  6. 2024PIXAR: Auto-Regressive Language Modeling in Pixel Spacecited by 16
  7. 2025Multilingual Pretraining for Pixel Language Modelscited by 12
  8. 2024Autoregressive Pre-Training on Pixels and Textscited by 10
  9. 2024How do Multimodal Foundation Models Encode Text and Speech? An Analysis of Cross-Lingual and Cross-Modal Representationscited by 8
  10. 2025FLEXITOKENS: Flexible Tokenization for Evolving Language Modelscited by 8
  11. 2025CLINIC: Evaluating Multilingual Trustworthiness in Language Models for Healthcarecited by 7
  12. 2023PixT3: Pixel-based Table-To-Text Generationcited by 6
10 more
  1. 2024LogogramNLP: Comparing Visual and Textual Representations of Ancient Logographic Writing Systems for NLPcited by 6
  2. 2025Overcoming Vocabulary Constraints with Pixel-level Fallbackcited by 6
  3. 2025CoCo-CoLa: Evaluating and Improving Language Adherence in Multilingual LLMscited by 4
  4. 2025Emergent Semantics Beyond Token Embeddings: Transformer LMs with Frozen Visual Unicode Representationscited by 3
  5. 2026Proxy Compression for Language Modelingcited by 3
  6. 2024Integrating Multi-scale Contextualized Information for Byte-based Neural Machine Translationcited by 1
  7. 2024Vision-fused Attack: Advancing Aggressive and Stealthy Adversarial Text against Neural Machine Translationcited by 1
  8. 2026Hot-Start Chinese Language Modeling:Visual Glyphs Accelerate Sample-Efficient Learningcited by 1
  9. 2026MIXAR: Scaling Autoregressive Pixel-based Language Models to Multiple Languages and Scriptscited by 1
  10. 2024MoCE: Adaptive Mixture of Contextualization Experts for Byte-based Neural Machine Translationcited by 0