# CRPO: Confidence-Reward Driven Preference Optimization for Machine Translation

arXiv: https://arxiv.org/abs/2501.13927 · HTML: https://mlsanity.com/paper/2501.13927/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in LLM preference optimization methods, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 23 of its 33 references that are in the map.
- First posted on arXiv 23 Jan 2025.

## Placed in

- LLM alignment / RLHF / preference optimisation / [LLM preference optimization methods](https://mlsanity.com/alignment/llm-preference-optimization-methods/)

## Builds on (most cited first, 23 of 23)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [No Language Left Behind: Scaling Human-Centered Machine Translation](https://mlsanity.com/paper/2207.04672/) (2022)
- [BLEURT: Learning Robust Metrics for Text Generation](https://mlsanity.com/paper/2004.04696/) (2020)
- [SimPO: Simple Preference Optimization with a Reference-Free Reward](https://mlsanity.com/paper/2405.14734/) (2024)
- [Reinforced Self-Training (ReST) for Language Modeling](https://mlsanity.com/paper/2308.08998/) (2023)
- [SLiC-HF: Sequence Likelihood Calibration with Human Feedback](https://mlsanity.com/paper/2305.10425/) (2023)
- [Contrastive Preference Optimization: Pushing the Boundaries of LLM Performance in Machine Translation](https://mlsanity.com/paper/2401.08417/) (2024)
- [Statistical Rejection Sampling Improves Preference Optimization](https://mlsanity.com/paper/2309.06657/) (2023)
- [xCOMET: Transparent Machine Translation Evaluation through Fine-grained Error Detection](https://mlsanity.com/paper/2310.10482/) (2023)
- [Calibrating Sequence likelihood Improves Conditional Language Generation](https://mlsanity.com/paper/2210.00045/) (2022)
- [Scaling up COMETKIWI: Unbabel-IST 2023 Submission for the Quality Estimation Shared Task](https://mlsanity.com/paper/2309.11925/) (2023)
- [Extrapolating Large Language Models to Non-English by Aligning Languages](https://mlsanity.com/paper/2308.04948/) (2023)
- [BigTranslate: Augmenting Large Language Models with Multilingual Translation Capability over 100 Languages](https://mlsanity.com/paper/2305.18098/) (2023)
- [TIM: Teaching Large Language Models to Translate with Comparison](https://mlsanity.com/paper/2307.04408/) (2023)
- [MBR and QE Finetuning: Training-time Distillation of the Best and Most Expensive Decoding Methods](https://mlsanity.com/paper/2309.10966/) (2023)
- [Improving Translation Faithfulness of Large Language Models via Augmenting Instructions](https://mlsanity.com/paper/2308.12674/) (2023)
- [Direct Preference Optimization for Neural Machine Translation with Minimum Bayes Risk Decoding](https://mlsanity.com/paper/2311.08380/) (2023)
- [Word Alignment as Preference for Machine Translation](https://mlsanity.com/paper/2405.09223/) (2024)
- [Advancing Translation Preference Modeling with RLHF: A Step Towards Cost-Effective Solution](https://mlsanity.com/paper/2402.11525/) (2024)

## Built on by (most cited first, 3 of 3)

- [On the Role of Preference Variance in Preference Optimization](https://mlsanity.com/paper/2510.13022/) (2025)
- [Cross-Preference Learning for Sentence-Level and Context-Aware Machine Translation](https://mlsanity.com/paper/2603.25183/) (2026)
- [STAR : Sentence Translation Alignment Rate for Document-to-Document Machine Translation](https://mlsanity.com/paper/2608.27161/) (2026)

## Abstract (arXiv)

Large language models (LLMs) have shown great potential in natural language processing tasks, but their application to machine translation (MT) remains challenging due to pretraining on English-centric data and the complexity of reinforcement learning from human feedback (RLHF). Direct Preference Optimization (DPO) has emerged as a simpler and more efficient alternative, but its performance depends heavily on the quality of preference data. To address this, we propose Confidence-Reward driven Preference Optimization (CRPO), a novel method that combines reward scores with model confidence to improve data selection for fine-tuning. CRPO selects challenging sentence pairs where the model is uncertain or underperforms, leading to more effective learning. While primarily designed for LLMs, CRPO also generalizes to encoder-decoder models like NLLB, demonstrating its versatility. Empirical results show that CRPO outperforms existing methods such as RS-DPO, RSO and MBR score in both translation accuracy and data efficiency.