mlsanityData to 14 Oct 2026

Unlocking Latent Discourse Translation in LLMs Through Quality-Aware Decoding

arXiv 2510.06866 ↗First posted 8 Oct 2025cs.CLcited by 0

Not on the map. Why?

Abstract (arXiv)

Large language models (LLMs) have emerged as strong contenders in machine translation.Yet, they still struggle to adequately handle discourse phenomena, such as pronoun resolution and lexical cohesion at the document level. In this study, we thoroughly investigate the discourse phenomena performance of LLMs in context-aware translation. We demonstrate that discourse knowledge is encoded within LLMs and propose the use of quality-aware decoding (QAD) to effectively extract this knowledge, showcasing its superiority over other decoding approaches through comprehensive analysis. Furthermore, we illustrate that QAD enhances the semantic richness of translations and aligns them more closely with human preferences.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 50 of its 82 references that are in the map. First posted on arXiv 8 Oct 2025.

Builds on

50 of 82 references.

References · 50

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 13,028
  3. 2022Large Language Models are Zero-Shot Reasonerscited by 4,374
  4. 2023Lost in the Middle: How Language Models Use Long Contextscited by 3,455
  5. 2014Sequence to Sequence Learning with Neural Networkscited by 2,986
  6. 2019The Curious Case of Neural Text Degenerationcited by 2,916
  7. 2018Hierarchical Neural Story Generationcited by 1,193
  8. 2020Making Monolingual Sentence Embeddings Multilingual using Knowledge Distillationcited by 593
  9. 2023xCOMET: Transparent Machine Translation Evaluation through Fine-grained Error Detectioncited by 265
  10. 2017Beam Search Strategies for Neural Machine Translationcited by 232
  11. 2024Tower: An Open Multilingual Large Language Model for Translation-Related Taskscited by 206
  12. 2020Is MAP Decoding All You Need? The Inadequacy of the Mode in Neural Machine Translationcited by 138
38 more
  1. 2024EuroLLM: Multilingual Language Models for Europecited by 106
  2. 2017Neural Machine Translation with Extended Contextcited by 95
  3. 2018Has Machine Translation Achieved Human Parity? A Case for Document-level Evaluationcited by 88
  4. 2017Evaluating Discourse Phenomena in Neural Machine Translationcited by 87
  5. 2018Document-Level Neural Machine Translation with Hierarchical Attention Networkscited by 85
  6. 2025WMT24++: Expanding the Language Coverage of WMT24 to 55 Languages & Dialectscited by 79
  7. 2018Improving the Transformer Translation Model with Document-Level Contextcited by 74
  8. 2021Sampling-Based Approximations to Minimum Bayes Risk Decoding for Neural Machine Translationcited by 65
  9. 2019Selective Attention for Context-aware Neural Machine Translationcited by 60
  10. 2024Adapting Large Language Models for Document-Level Machine Translationcited by 59
  11. 2021Understanding the Properties of Minimum Bayes Risk Decoding in Neural Machine Translationcited by 56
  12. 2017Exploiting Cross-Sentence Context for Neural Machine Translationcited by 48
  13. 2020Energy-Based Reranking: Improving Neural Machine Translation Using Energy-Based Modelscited by 44
  14. 2025M-Prometheus: A Suite of Open Multilingual LLM Judgescited by 38
  15. 2021When Does Translation Require Context? A Data-driven, Multilingual Explorationcited by 36
  16. 2022DiscoScore: Evaluating Text Generation with BERT and Discourse Coherencecited by 36
  17. 2024Gender Bias in Large Language Models across Multiple Languagescited by 35
  18. 2020Does Multi-Encoder Help? A Case Study on Context-Aware Neural Machine Translationcited by 34
  19. 2017Attention Strategies for Multi-Source Sequence-to-Sequence Learningcited by 32
  20. 2023Do GPTs Produce Less Literal Translations?cited by 30
  21. 2023Escaping the sentence-level paradigm in machine translationcited by 27
  22. 2024DelTA: An Online Document-Level Translation Agent Based on Multi-Level Memorycited by 27
  23. 2020Diving Deep into Context-Aware Neural Machine Translationcited by 25
  24. 2024Efficiently Exploring Large Language Models for Document-Level Machine Translation with In-context Learningcited by 23
  25. 2024Guardians of the Machine Translation Meta-Evaluation: Sentinel Metrics Fall In!cited by 21
  26. 2024xTower: A Multilingual LLM for Explaining and Correcting Translation Errorscited by 17
  27. 2024Don't Rank, Combine! Combining Machine Translation Hypotheses Using Quality Estimationcited by 13
  28. 2024Enhancing Document-level Translation of Large Language Model via Translation Mixed-instructionscited by 13
  29. 2021Divide and Rule: Effective Pre-Training for Context-Aware Multi-Encoder Translation Modelscited by 12
  30. 2023Document-Level Language Models for Machine Translationcited by 12
  31. 2025Multilingual Contextualization of Large Language Models for Document-Level Machine Translationcited by 11
  32. 2019Using Whole Document Context in Neural Machine Translationcited by 9
  33. 2024On Measuring Context Utilization in Document-Level MT Systemscited by 9
  34. 2020Can Your Context-Aware MT System Pass the DiP Benchmark Tests? : Evaluation Benchmarks for Discourse Phenomena in Machine Translationcited by 6
  35. 2024Context-aware and Style-related Incremental Decoding framework for Discourse-Level Literary Translationcited by 5
  36. 2024Choose the Final Translation from NMT and LLM hypotheses Using MBR Decoding: HW-TSC's Submission to the WMT24 General MT Shared Taskcited by 4
  37. 2023Context-aware Neural Machine Translation for English-Japanese Business Scene Dialoguescited by 1
  38. 2024FAME-MT Dataset: Formality Awareness Made Easy for Machine Translation Purposescited by 1

Built on by

Not cited yet.