Saved in:
Bibliographic Details
Main Authors: Tan, Shaomu, Mitani, Ryosuke, Choudhary, Ritvik, Sekiya, Toshiyuki
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2509.19020
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909802458775552
author Tan, Shaomu
Mitani, Ryosuke
Choudhary, Ritvik
Sekiya, Toshiyuki
author_facet Tan, Shaomu
Mitani, Ryosuke
Choudhary, Ritvik
Sekiya, Toshiyuki
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2509_19020
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Investigating Test-Time Scaling with Reranking for Machine Translation
Tan, Shaomu
Mitani, Ryosuke
Choudhary, Ritvik
Sekiya, Toshiyuki
Computation and Language
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.
title Investigating Test-Time Scaling with Reranking for Machine Translation
topic Computation and Language
url https://arxiv.org/abs/2509.19020