mlsanityData to 7 Oct 2026

STAR : Sentence Translation Alignment Rate for Document-to-Document Machine Translation

arXiv 2608.27161 ↗First posted 27 Aug 2026cs.CLcited by 1

On the map: LLM alignment / RLHF / preference optimisation → LLM preference optimization methods

Abstract (arXiv)

Large Language Models (LLMs) have enabled a shift from sentence-level to document-to-document (Doc2Doc) machine translation, promising improved global coherence. However, document-to-document generation in a single pass frequently suffers from structural misalignment, manifesting as sentence omissions or hallucinations that violate the core requirement of source-target correspondence. To address this, we introduce Sentence Translation Alignment Rate (STAR), an auxiliary metric that explicitly quantifies sentence-level structural fidelity. Building on this, we propose STAR-masked Preference Optimization (StarPO), a framework that ranks document-level hypotheses by structural quality and utilizes a dynamic alignment mask to focus optimization on misaligned segments. Experimental results across news and literary domains demonstrate that StarPO significantly enhances translation quality and structural integrity. Notably, StarPO allows compact models to surpass the performance of massive proprietary systems like GPT-4o while maintaining superior token efficiency.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Placed in LLM preference optimization methods, one of 10 directions in LLM alignment / RLHF / preference optimisation. Builds on 61 of its 84 references that are in the map. First posted on arXiv 27 Aug 2026.

Builds on

61 of 84 references.

Closest ancestors · 16

Same direction, newest first

  1. 2025M2PO: Multi-Perspective Multi-Pair Preference Optimization for Machine Translation15 Oct
  2. 2025Align-then-Slide: A complete evaluation framework for Ultra-Long Document-Level Machine Translation4 Sep
  3. 2025RIVAL: Reinforcement Learning with Iterative and Adversarial Optimization for Machine Translation5 Jun
  4. 2025Remedy: Learning Machine Translation Evaluation from Human Preferences with Reward Modeling18 Apr
  5. 2025Mask-DPO: Generalizable Fine-grained Factuality Alignment of LLMs4 Mar
  6. 2025CRPO: Confidence-Reward Driven Preference Optimization for Machine Translation23 Jan
  7. 2024Fine-Grained Reward Optimization for Machine Translation using Error Severity Mappings8 Nov
  8. 2024Modeling User Preferences with Automatic Metrics: Creating a High-Quality Preference Dataset for Machine Translation10 Oct
  9. 2024SimPO: Simple Preference Optimization with a Reference-Free Reward23 May
  10. 2024Word Alignment as Preference for Machine Translation15 May
  11. 2024ORPO: Monolithic Preference Optimization without Reference Model12 Mar
  12. 2024KTO: Model Alignment as Prospect Theoretic Optimization2 Feb
4 more
  1. 2024Improving Machine Translation with Human Feedback: An Exploration of Quality Estimation as a Reward Model23 Jan
  2. 2024Contrastive Preference Optimization: Pushing the Boundaries of LLM Performance in Machine Translation16 Jan
  3. 2023Direct Preference Optimization for Neural Machine Translation with Minimum Bayes Risk Decoding14 Nov
  4. 2022Training language models to follow instructions with human feedback4 Mar

Other references · 45

  1. 2024The Llama 3 Herd of Modelscited by 14,192
  2. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,084
  3. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 8,851
  4. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 7,916
  5. 2023Lost in the Middle: How Language Models Use Long Contextscited by 3,383
  6. 2024M3-Embedding: Multi-Linguality, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillationcited by 1,050
  7. 2024SWIFT:A Scalable lightWeight Infrastructure for Fine-Tuningcited by 345
  8. 2023Prompting Large Language Model for Machine Translation: A Case Studycited by 247
  9. 2024Tower: An Open Multilingual Large Language Model for Translation-Related Taskscited by 205
  10. 2024Hunyuan-Large: An Open-Source MoE Model with 52 Billion Activated Parameters by Tencentcited by 89
  11. 2023Large language models effectively leverage document-level context for literary translation, but critical errors persistcited by 80
  12. 2024Adapting Large Language Models for Document-Level Machine Translationcited by 59
33 more
  1. 2022Embarrassingly Easy Document-Level MT Metrics: How to Convert Any Pretrained Metric Into a Document-Level Metriccited by 56
  2. 2023TIM: Teaching Large Language Models to Translate with Comparisoncited by 50
  3. 2025MT-R1-Zero: Advancing LLM-based Machine Translation via R1-Zero-like Reinforcement Learningcited by 48
  4. 2024Segment Any Text: A Universal Approach for Robust, Efficient and Adaptable Sentence Segmentationcited by 41
  5. 2025Tower+: Bridging Generality and Translation Specialization in Multilingual LLMscited by 40
  6. 2024X-ALMA: Plug & Play Modules and Adaptive Rejection for Quality Translation at Scalecited by 36
  7. 2024Translating Step-by-Step: Decomposing the Translation Process for Improved Translation Quality of Long-Form Textscited by 31
  8. 2025Beyond the Surface: Measuring Self-Preference in LLM Judgmentscited by 30
  9. 2024DRT: Deep Reasoning Translation via Long Chain-of-Thoughtcited by 28
  10. 2024DelTA: An Online Document-Level Translation Agent Based on Multi-Level Memorycited by 27
  11. 2024Efficiently Exploring Large Language Models for Document-Level Machine Translation with In-context Learningcited by 23
  12. 2024The Fine-Tuning Paradox: Boosting Translation Quality Without Sacrificing LLM Abilitiescited by 22
  13. 2023Contextual Refinement of Translations: Large Language Models for Sentence and Document-Level Post-Editingcited by 21
  14. 2024In-Context Example Selection via Similarity Search Improves Low-Resource Machine Translationcited by 21
  15. 2025Detecting Proxy Gaming in RL and LLM Alignment via Evaluator Stress Testscited by 19
  16. 2024Salute the Classic: Revisiting Challenges of Machine Translation in the Age of Large Language Modelscited by 17
  17. 2024Enhancing Document-level Translation of Large Language Model via Translation Mixed-instructionscited by 13
  18. 2025ExTrans: Multilingual Deep Reasoning Translation via Exemplar-Enhanced Reinforcement Learningcited by 11
  19. 2023WSPAlign: Word Alignment Pre-training via Large-Scale Weakly Supervised Span Predictioncited by 10
  20. 2024Investigating Length Issues in Document-level Machine Translationcited by 10
  21. 2023SentAlign: Accurate and Scalable Sentence Alignmentcited by 9
  22. 2025MT-RewardTree: A Comprehensive Framework for Advancing LLM-Based Machine Translation via Reward Modelingcited by 9
  23. 2025Source-primed Multi-turn Conversation Helps Large Language Models Translate Documentscited by 7
  24. 2024DeMPT: Decoding-enhanced Multi-phase Prompt Tuning for Making LLMs Be Better Context-aware Translatorscited by 6
  25. 2025Doc-Guided Sent2Sent++: A Sent2Sent++ Agent with Doc-Guided memory for Document-level Machine Translationcited by 6
  26. 2025Mitigating Hallucinated Translations in Large Language Models with Hallucination-focused Preference Optimizationcited by 6
  27. 2025Extending Automatic Machine Translation Evaluation to Book-Length Documentscited by 6
  28. 2024Context-Aware or Context-Insensitive? Assessing LLMs' Performance in Document-Level Translationcited by 5
  29. 2025Two Intermediate Translations Are Better Than One: Fine-tuning LLMs for Document-level Translation Refinementcited by 5
  30. 2024Analyzing the Attention Heads for Pronoun Disambiguation in Context-aware Machine Translation Modelscited by 4
  31. 2025You Are What You Train: Effects of Data Composition on Training Context-aware Machine Translation Modelscited by 3
  32. 2026Cross-Preference Learning for Sentence-Level and Context-Aware Machine Translationcited by 2
  33. 2024Understanding and Addressing the Under-Translation Problem from the Perspective of Decoding Objectivecited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026When Do Agents Help? Embedding, LLM and Agentic Alignment of Classical Texts and Their Translationscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Paraphrase-Aligned Machine Translation
  2. 2025STARS: Synchronous Token Alignment for Robust Supervision in Large Language Models
  3. 2024Is Preference Alignment Always the Best Option to Enhance LLM-Based Translation? An Empirical Analysis
  4. 2026Enhancing Document-Level Machine Translation via Filtered Synthetic Corpora and Two-Stage LLM Adaptation
  5. 2019Jointly Learning to Align and Translate with Transformer Models
  6. 2025DiffPO: Diffusion-styled Preference Optimization for Efficient Inference-Time Alignment of Large Language Models
  7. 2025TransAlign: Machine Translation Encoders are Strong Word Aligners, Too
  8. 2025Bridging the Linguistic Divide: A Survey on Leveraging Large Language Models for Machine Translation