Investigating Test-Time Scaling with Reranking for Machine Translation
On the map: LLM reasoning → Test-time compute and efficient thinking
Abstract (arXiv)
Scaling model parameters has become the de facto strategy for improving NLP systems, but it comes with substantial computational costs. Test-Time Scaling (TTS) offers an alternative by allocating more computation at inference: generating multiple candidates and selecting the best. While effective in tasks such as mathematical reasoning, TTS has not been systematically explored for machine translation (MT). In this paper, we present the first systematic study of TTS for MT, investigating a simple but practical best-of-N framework on WMT24 benchmarks. Our experiments cover six high-resource and one low-resource language pairs, five model sizes (3B-72B), and various TTS compute budget (N up to 1024). Our results show that a) For high-resource languages, TTS generally improves translation quality according to multiple neural MT evaluation metrics, and our human evaluation confirms these gains; b) Augmenting smaller models with large N can match or surpass larger models at N=1 with more compute cost; c) Under fixed compute budgets, larger models are typically more efficient, and TTS can degrade quality due to metric blind spots in low-resource cases.
Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning. Builds on 13 of its 39 references that are in the map. First posted on arXiv 23 Sep 2025.
Builds on
13 of 39 references.
Closest ancestors · 2
Same direction, newest first
- 2025s1: Simple test-time scaling31 Jan
- 2024Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters6 Aug
Other references · 11
- 2023GPT-4 Technical Reportcited by 18,396
- 2020Scaling Laws for Neural Language Modelscited by 6,586
- 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
- 2024Qwen2.5-Coder Technical Reportcited by 1,203
- 2017Deep Learning Scaling is Predictable, Empiricallycited by 582
- 2024Tower: An Open Multilingual Large Language Model for Translation-Related Taskscited by 206
- 2022Quality-Aware Decoding for Neural Machine Translationcited by 49
- 2025MT-R1-Zero: Advancing LLM-based Machine Translation via R1-Zero-like Reinforcement Learningcited by 48
- 2025Remedy: Learning Machine Translation Evaluation from Human Preferences with Reward Modelingcited by 9
- 2025Please Translate Again: Two Simple Experiments on Whether Human-Like Reasoning Helps Translationcited by 7
- 2025Adding Chocolate to Mint: Mitigating Metric Interference in Machine Translationcited by 6
Built on by
Citing papers per year: 2025: 2; 2026: 1.
3 papers.
Similar, not cited
Close in wording, no citation either way
- 2026Ladders in Chaos: When, How, (and Perhaps Why) Does Test-Time Scaling Improve LLM Machine Translation
- 2025The Art of Scaling Test-Time Compute for Large Language Models
- 2024Inference Scaling Laws: An Empirical Analysis of Compute-Optimal Inference for Problem-Solving with Language Models
- 2025A Survey on Test-Time Scaling in Large Language Models: What, How, Where, and How Well?
- 2026Recovered in Translation: Efficient Pipeline for Automated Translation of Benchmarks and Datasets
- 2025Test-time Scaling of LLMs: A Survey from A Subproblem Structure Perspective
- 2025AgentTTS: Large Language Model Agent for Test-time Compute-optimal Scaling Strategy in Complex Tasks
- 2025Towards Sustainable NLP: Insights from Benchmarking Inference Energy in Large Language Models