# Dynamic Context Selection for Document-level Neural Machine Translation via Reinforcement Learning

arXiv: https://arxiv.org/abs/2010.04314 · HTML: https://mlsanity.com/paper/2010.04314/

## Key facts

- Cited by 25 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 9 Oct 2020.

## Built on by (most cited first, 25 of 25)

- [Large language models effectively leverage document-level context for literary translation, but critical errors persist](https://mlsanity.com/paper/2304.03245/) (2023)
- [Scientific Credibility of Machine Translation Research: A Meta-Evaluation of 769 Papers](https://mlsanity.com/paper/2106.15195/) (2021)
- [Modeling Bilingual Conversational Characteristics for Neural Chat Translation](https://mlsanity.com/paper/2107.11164/) (2021)
- [Divide and Rule: Effective Pre-Training for Context-Aware Multi-Encoder Translation Models](https://mlsanity.com/paper/2103.17151/) (2021)
- [KGPool: Dynamic Knowledge Graph Context Selection for Relation Extraction](https://mlsanity.com/paper/2106.00459/) (2021)
- [Towards Making the Most of Dialogue Characteristics for Neural Chat Translation](https://mlsanity.com/paper/2109.00668/) (2021)
- [Exploring Dynamic Selection of Branch Expansion Orders for Code Generation](https://mlsanity.com/paper/2106.00261/) (2021)
- [Investigating Length Issues in Document-level Machine Translation](https://mlsanity.com/paper/2412.17592/) (2024)
- [Multilingual Document-Level Translation Enables Zero-Shot Transfer From Sentences to Documents](https://mlsanity.com/paper/2109.10341/) (2021)
- [Focused Concatenation for Context-Aware Neural Machine Translation](https://mlsanity.com/paper/2210.13388/) (2022)
- [Encoding Sentence Position in Context-Aware Neural Machine Translation with Concatenation](https://mlsanity.com/paper/2302.06459/) (2023)
- [WISK: A Workload-aware Learned Index for Spatial Keyword Queries](https://mlsanity.com/paper/2302.14287/) (2023)
- [DeMPT: Decoding-enhanced Multi-phase Prompt Tuning for Making LLMs Be Better Context-aware Translators](https://mlsanity.com/paper/2402.15200/) (2024)
- [P-Transformer: Towards Better Document-to-Document Neural Machine Translation](https://mlsanity.com/paper/2212.05830/) (2022)
- [Self-Modifying State Modeling for Simultaneous Machine Translation](https://mlsanity.com/paper/2406.02237/) (2024)
- [Towards Personalised and Document-level Machine Translation of Dialogue](https://mlsanity.com/paper/2102.10979/) (2021)
- [Learning as Conversation: Dialogue Systems Reinforced for Information Acquisition](https://mlsanity.com/paper/2205.14748/) (2022)
- [A Case Study on Context-Aware Neural Machine Translation with Multi-Task Learning](https://mlsanity.com/paper/2407.03076/) (2024)
- [A Multi-task Multi-stage Transitional Training Framework for Neural Chat Translation](https://mlsanity.com/paper/2301.11749/) (2023)
- [Exploring Paracrawl for Document-level Neural Machine Translation](https://mlsanity.com/paper/2304.10216/) (2023)
- [SAGE: Sustainable Agent-Guided Expert-tuning for Culturally Attuned Translation in Low-Resource Southeast Asia](https://mlsanity.com/paper/2603.19931/) (2026)
- [A Case Study on Context Encoding in Multi-Encoder based Document-Level Neural Machine Translation](https://mlsanity.com/paper/2308.06063/) (2023)
- [Token-level Accept or Reject: A Micro Alignment Approach for Large Language Models](https://mlsanity.com/paper/2505.19743/) (2025)
- [Context-Adaptive Document-Level Neural Machine Translation](https://mlsanity.com/paper/2104.08259/) (2021)
- [Interpretable Style Takagi-Sugeno-Kang Fuzzy Clustering](https://mlsanity.com/paper/2504.05125/) (2025)

## Abstract (arXiv)

Document-level neural machine translation has yielded attractive improvements. However, majority of existing methods roughly use all context sentences in a fixed scope. They neglect the fact that different source sentences need different sizes of context. To address this problem, we propose an effective approach to select dynamic context so that the document-level translation model can utilize the more useful selected context sentences to produce better translations. Specifically, we introduce a selection module that is independent of the translation module to score each candidate context sentence. Then, we propose two strategies to explicitly select a variable number of context sentences and feed them into the translation module. We train the two modules end-to-end via reinforcement learning. A novel reward is proposed to encourage the selection and utilization of dynamic context sentences. Experiments demonstrate that our approach can select adaptive context sentences for different source sentences, and significantly improves the performance of document-level translation methods.