# TranslateGemma Technical Report

arXiv: https://arxiv.org/abs/2601.09012 · HTML: https://mlsanity.com/paper/2601.09012/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 7 of its 14 references that are in the map.
- First posted on arXiv 13 Jan 2026.

## Builds on (most cited first, 7 of 7)

- [Adafactor: Adaptive Learning Rates with Sublinear Memory Cost](https://mlsanity.com/paper/1804.04235/) (2018)
- [Experts, Errors, and Context: A Large-Scale Study of Human Evaluation for Machine Translation](https://mlsanity.com/paper/2104.14478/) (2021)
- [MADLAD-400: A Multilingual And Document-Level Large Audited Dataset](https://mlsanity.com/paper/2309.04662/) (2023)
- [MetricX-24: The Google Submission to the WMT 2024 Metrics Shared Task](https://mlsanity.com/paper/2410.03983/) (2024)
- [WMT24++: Expanding the Language Coverage of WMT24 to 55 Languages & Dialects](https://mlsanity.com/paper/2502.12404/) (2025)
- [Finding Replicable Human Evaluations via Stable Ranking Probability](https://mlsanity.com/paper/2404.01474/) (2024)
- [Introducing the NewsPaLM MBR and QE Dataset: LLM-Generated High-Quality Parallel Data Outperforms Traditional Web-Crawled Data](https://mlsanity.com/paper/2408.06537/) (2024)

## Built on by (most cited first, 7 of 7)

- [Efficient Multilingual Reasoning Transfer via Progressive Code-Switching](https://mlsanity.com/paper/2607.00485/) (2026)
- [Consensus Group Relative Policy Optimization for Text Generation](https://mlsanity.com/paper/2602.03102/) (2026)
- [AMALIA-VL: A Native European Portuguese Open-Source Vision and Language Model](https://mlsanity.com/paper/2606.19100/) (2026)
- [Reasoning Before Translation: Enhancing Legal Machine Translation with Structured Reasoning](https://mlsanity.com/paper/2607.19181/) (2026)
- [Semantic Alignment of AI Models: Concept Collapse, Checkpoint Dynamics, and Cross-Lingual Transfer](https://mlsanity.com/paper/2608.01585/) (2026)
- [TranslatePsy-AfriSLM: High-Quality Data Scaling For Low-Resource Machine Translation](https://mlsanity.com/paper/2608.18655/) (2026)
- [In the Blind: Building Pseudo-References for MT Evaluation](https://mlsanity.com/paper/2609.13611/) (2026)

## Abstract (arXiv)

We present TranslateGemma, a suite of open machine translation models based on the Gemma 3 foundation models. To enhance the inherent multilingual capabilities of Gemma 3 for the translation task, we employ a two-stage fine-tuning process. First, supervised fine-tuning is performed using a rich mixture of high-quality large-scale synthetic parallel data generated via state-of-the-art models and human-translated parallel data. This is followed by a reinforcement learning phase, where we optimize translation quality using an ensemble of reward models, including MetricX-QE and AutoMQM, targeting translation quality. We demonstrate the effectiveness of TranslateGemma with human evaluation on the WMT25 test set across 10 language pairs and with automatic evaluation on the WMT24++ benchmark across 55 language pairs. Automatic metrics show consistent and substantial gains over the baseline Gemma 3 models across all sizes. Notably, smaller TranslateGemma models often achieve performance comparable to larger baseline models, offering improved efficiency. We also show that TranslateGemma models retain strong multimodal capabilities, with enhanced performance on the Vistra image translation benchmark. The release of the open TranslateGemma models aims to provide the research community with powerful and adaptable tools for machine translation.