mlsanityData to 7 Oct 2026

Understanding the Properties of Minimum Bayes Risk Decoding in Neural Machine Translation

arXiv 2105.08504 ↗First posted 18 May 2021cs.CL cs.LGcited by 56

Not on the map. Why?

Abstract (arXiv)

Neural Machine Translation (NMT) currently exhibits biases such as producing translations that are too short and overgenerating frequent words, and shows poor robustness to copy noise in training data or domain shift. Recent work has tied these shortcomings to beam search -- the de facto standard inference algorithm in NMT -- and Eikema & Aziz (2020) propose to use Minimum Bayes Risk (MBR) decoding on unbiased samples instead. In this paper, we empirically investigate the properties of MBR decoding on a number of previously reported biases and failure cases of beam search. We find that MBR still exhibits a length and token frequency bias, owing to the MT metrics used as utility functions, but that MBR also increases robustness against copy noise in the training data and domain shift.

Key facts. Cited by 56 papers in the map (data through 7 Oct 2026). Builds on 23 of its 40 references that are in the map. First posted on arXiv 18 May 2021.

Builds on

23 of 40 references.

References · 23

  1. 2017Attention Is All You Needcited by 59,932
  2. 2014Sequence to Sequence Learning with Neural Networkscited by 2,974
  3. 2018A Call for Clarity in Reporting BLEU Scorescited by 1,790
  4. 2016Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translationcited by 1,471
  5. 2020COMET: A Neural Framework for MT Evaluationcited by 1,024
  6. 2015Scheduled Sampling for Sequence Prediction with Recurrent Neural Networkscited by 779
  7. 2018Subword Regularization: Improving Neural Network Translation Models with Multiple Subword Candidatescited by 632
  8. 2017Six Challenges for Neural Machine Translationcited by 334
  9. 2020Is MAP Decoding All You Need? The Inadequacy of the Mode in Neural Machine Translationcited by 137
  10. 2020The Tatoeba Translation Challenge -- Realistic Data Sets for Low Resource and Multilingual MTcited by 127
  11. 2016Sequence-to-Sequence Learning as Beam-Search Optimizationcited by 126
  12. 2018Analyzing Uncertainty in Neural Machine Translationcited by 126
11 more
  1. 2019Bridging the Gap between Training and Inference for Neural Machine Translationcited by 123
  2. 2015Minimum Risk Training for Neural Machine Translationcited by 114
  3. 2019On NMT Search Errors and Model Errors: Cat Got Your Tongue?cited by 114
  4. 2020SSMBA: Self-Supervised Manifold Based Data Augmentation for Improving Out-of-Domain Robustnesscited by 94
  5. 2018On the Impact of Various Types of Noise on Neural Machine Translationcited by 76
  6. 2019Calibration of Encoder Decoder Models for Neural Machine Translationcited by 67
  7. 2019Domain Robustness in Neural Machine Translationcited by 64
  8. 2018Dual Conditional Cross-Entropy Filtering of Noisy Parallel Corporacited by 40
  9. 2016Neural Machine Translation by Minimising the Bayes-risk with Respect to Syntactic Translation Latticescited by 17
  10. 2020The Sockeye 2 Neural Machine Translation Toolkit at AMTA 2020cited by 16
  11. 2017Later-stage Minimum Bayes-Risk Decoding for Neural Machine Translationcited by 8

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 4; 2022: 8; 2023: 16; 2024: 17; 2025: 5; 2026: 6.

56 papers.

Sort:

By area:

  1. 2024Next-Generation Database Interfaces: A Survey of LLM-based Text-to-SQLcited by 118
  2. 2021High Quality Rather than High Model Probability: Minimum Bayes Risk Decoding with Neural Metricscited by 104
  3. 2022Detecting and Mitigating Hallucinations in Machine Translation: Model Internal Workings Alone Do Well, Sentence Similarity Even Bettercited by 75
  4. 2021Sampling-Based Approximations to Minimum Bayes Risk Decoding for Neural Machine Translationcited by 63
  5. 2022Looking for a Needle in a Haystack: A Comprehensive Study of Hallucinations in Neural Machine Translationcited by 60
  6. 2023Epsilon Sampling Rocks: Investigating Sampling Strategies for Minimum Bayes Risk Decoding for Machine Translationcited by 57
  7. 2022Quality-Aware Decoding for Neural Machine Translationcited by 49
  8. 2023It's MBR All the Way Down: Modern Generation Techniques Through the Lens of Minimum Bayes Riskcited by 49
  9. 2024Bidirectional Decoding: Improving Action Chunking via Guided Test-Time Samplingcited by 46
  10. 2023Exploring Human-Like Translation Strategy with Large Language Modelscited by 44
  11. 2023DelucionQA: Detecting Hallucinations in Domain-specific Question Answeringcited by 43
  12. 2023Mitigating Hallucinations and Off-target Machine Translation with Source-Contrastive and Language-Contrastive Decodingcited by 42
38 more
  1. 2022Identifying Weaknesses in Machine Translation Metrics Through Minimum Bayes Risk Decoding: A Case Study for COMETcited by 41
  2. 2023HalOmi: A Manually Annotated Benchmark for Multilingual Hallucination and Omission Detection in Machine Translationcited by 35
  3. 2021Why don't people use character-level machine translation?cited by 27
  4. 2023BLEURT Has Universal Translations: An Analysis of Automatic Metrics by Minimum Risk Trainingcited by 24
  5. 2023Aligning Neural Machine Translation Models: Human Feedback in Training and Inferencecited by 19
  6. 2024Hyperparameter-Free Approach for Faster Minimum Bayes Risk Decodingcited by 15
  7. 2022Optimal Transport for Unsupervised Hallucination Detection in Neural Machine Translationcited by 13
  8. 2024Efficient Minimum Bayes Risk Decoding using Low-Rank Matrix Completion Algorithmscited by 13
  9. 2024Improving Minimum Bayes Risk Decoding with Multi-Promptcited by 13
  10. 2024mbrs: A Library for Minimum Bayes Risk Decodingcited by 13
  11. 2024Centroid-Based Efficient Minimum Bayes Risk Decodingcited by 12
  12. 2024Generating Diverse and High-Quality Texts by Minimum Bayes Risk Decodingcited by 11
  13. 2021Characterizing and addressing the issue of oversmoothing in neural autoregressive sequence modelingcited by 10
  14. 2023R-U-SURE? Uncertainty-Aware Code Suggestions By Maximizing Utility Across Random User Intentscited by 10
  15. 2023Model-Based Minimum Bayes Risk Decoding for Text Generationcited by 10
  16. 2024Introducing the NewsPaLM MBR and QE Dataset: LLM-Generated High-Quality Parallel Data Outperforms Traditional Web-Crawled Datacited by 10
  17. 2022Uncertainty Determines the Adequacy of the Mode and the Tractability of Decoding in Sequence-to-Sequence Modelscited by 9
  18. 2023Quality-Aware Translation Models: Efficient Generation and Quality Estimation in a Single Modelcited by 9
  19. 2024Structure Guided Large Language Model for SQL Generationcited by 9
  20. 2024Large Language Models for Dysfluency Detection in Stuttered Speechcited by 8
  21. 2024Diversity Explains Inference Scaling Laws: Through a Case Study of Minimum Bayes Risk Decodingcited by 8
  22. 2023Breeding Machine Translations: Evolutionary approach to survive and thrive in the world of automated evaluationcited by 5
  23. 2023Investigating Multi-Pivot Ensembling with Massively Multilingual Machine Translation Modelscited by 5
  24. 2022DC-MBR: Distributional Cooling for Minimum Bayesian Risk Decodingcited by 4
  25. 2023Minimum Bayes' Risk Decoding for System Combination of Grammatical Error Correction Systemscited by 4
  26. 2023Progressive Translation: Improving Domain Robustness of Neural Machine Translation with Intermediate Sequencescited by 3
  27. 2025Theoretical Guarantees for Minimum Bayes Risk Decodingcited by 3
  28. 2022CUNI Submission in WMT22 General Taskcited by 2
  29. 2024On the True Distribution Approximation of Minimum Bayes-Risk Decodingcited by 2
  30. 2026Task-Awareness Improves LLM Generations and Uncertaintycited by 2
  31. 2024EBBS: An Ensemble with Bi-Level Beam Search for Zero-Shot Machine Translationcited by 1
  32. 2024OTTAWA: Optimal TransporT Adaptive Word Aligner for Hallucination and Omission Translation Errors Detectioncited by 1
  33. 2025Quality Estimation Reranking for Document-Level Translationcited by 1
  34. 2023O-1: Self-training with Oracle and 1-best Hypothesiscited by 0
  35. 2024Non-Fluent Synthetic Target-Language Data Improve Neural Machine Translationcited by 0
  36. 2024Thesis proposal: Are We Losing Textual Diversity to Natural Language Processing?cited by 0
  37. 2025Multi-Hypothesis Distillation of Multilingual Neural Translation Models for Low-Resource Languagescited by 0
  38. 2025Unlocking Latent Discourse Translation in LLMs Through Quality-Aware Decodingcited by 0

50 most cited of 56 All 56 (JSON)