Translate-R1: Cost-Aware Translation Tool Use via Reinforcement Learning
Quick Answer
Translate-R1 introduces a cost-aware translation tool using reinforcement learning, enhancing Qwen3-4B's performance across 22 languages.
Quick Take
The gated policy improves reward by +4.6 on High, +23.5 on Low, and +17.5 on XLow resource tiers, achieving 63% cost efficiency compared to unconstrained translation.
Key Points
- Developed a single policy for translation decision-making based on reward feedback.
- Achieved Pareto-optimal performance across 87% of cost-sensitivity range.
- Improved performance on synthetic languages by +18.7 over the baseline.
- The policy transfers effectively to 9 held-out languages without prior training.
- Utilized a data pipeline that preserves answers while translating.
Paper Resources
Source Excerpt
arXiv:2606. 06835v1 Announce Type: new Abstract: The performance gap across languages in is well documented, and closing it natively requires pretraining or fine-tuning on corpora that, for most languages, do not exist. Translation offers an alternative: converting an input into the model's dominant language unlocks its full capabilities at once.
Applying translation to every input, however, is wasteful for languages the model already handles, while leaving the choice to the model fails in the opposite way, as LLMs are overconfident and skip the tool even when they cannot understand the input. …
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CL
See more →TriAgent: Divergence-Aware Committees for Cost-Efficient Financial Sentiment Analysis
TriAgent introduces a cost-efficient multi-agent system for financial sentiment analysis, combining VADER, FinBERT, and Qwen2.5. It achieves an F1 score of ~0.87 with significant savings of $9.3M/year at a 10M-user scale compared to GPT-4o-mini, while also detecting hallucinations with an AUC of 0.90.