# Alibaba-Translate China's Submission for WMT 2022 Metrics Shared Task

arXiv: https://arxiv.org/abs/2210.09683 · HTML: https://mlsanity.com/paper/2210.09683/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 16 of its 34 references that are in the map.
- First posted on arXiv 18 Oct 2022.

## Builds on (most cited first, 16 of 16)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [BERTScore: Evaluating Text Generation with BERT](https://mlsanity.com/paper/1904.09675/) (2019)
- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [BLEURT: Learning Robust Metrics for Text Generation](https://mlsanity.com/paper/2004.04696/) (2020)
- [COMET: A Neural Framework for MT Evaluation](https://mlsanity.com/paper/2009.09025/) (2020)
- [Experts, Errors, and Context: A Large-Scale Study of Human Evaluation for Machine Translation](https://mlsanity.com/paper/2104.14478/) (2021)
- [InfoXLM: An Information-Theoretic Framework for Cross-Lingual Language Model Pre-Training](https://mlsanity.com/paper/2007.07834/) (2020)
- [TransQuest: Translation Quality Estimation with Cross-lingual Transformers](https://mlsanity.com/paper/2011.01536/) (2020)
- [Uncertainty-Aware Machine Translation Evaluation](https://mlsanity.com/paper/2109.06352/) (2021)
- [UniTE: Unified Translation Evaluation](https://mlsanity.com/paper/2204.13346/) (2022)
- [Self-Paced Learning for Neural Machine Translation](https://mlsanity.com/paper/2010.04505/) (2020)
- [Learning to Generalize to More: Continuous Semantic Augmentation for Neural Machine Translation](https://mlsanity.com/paper/2204.06812/) (2022)
- [Towards User-Driven Neural Machine Translation](https://mlsanity.com/paper/2106.06200/) (2021)
- [RoBLEURT Submission for the WMT2021 Metrics Task](https://mlsanity.com/paper/2204.13352/) (2022)
- [KGR^4: Retrieval, Retrospect, Refine and Rethink for Commonsense Generation](https://mlsanity.com/paper/2112.08266/) (2021)

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

- [GEMBA-MQM: Detecting Translation Quality Error Spans with GPT-4](https://mlsanity.com/paper/2310.13988/) (2023)
- [MetricX-24: The Google Submission to the WMT 2024 Metrics Shared Task](https://mlsanity.com/paper/2410.03983/) (2024)
- [ACES: Translation Accuracy Challenge Sets for Evaluating Machine Translation Metrics](https://mlsanity.com/paper/2210.15615/) (2022)
- [Has Machine Translation Evaluation Achieved Human Parity? The Human Reference and the Limits of Progress](https://mlsanity.com/paper/2506.19571/) (2025)
- [Multi-task Adversarial Attacks against Black-box Model with Few-shot Queries](https://mlsanity.com/paper/2508.10039/) (2025)
- [Towards Multiple References Era -- Addressing Data Leakage and Limited Reference Diversity in NLG Evaluation](https://mlsanity.com/paper/2308.03131/) (2023)
- [Evaluating Automatic Metrics with Incremental Machine Translation Systems](https://mlsanity.com/paper/2407.03277/) (2024)

## Abstract (arXiv)

In this report, we present our submission to the WMT 2022 Metrics Shared Task. We build our system based on the core idea of UNITE (Unified Translation Evaluation), which unifies source-only, reference-only, and source-reference-combined evaluation scenarios into one single model. Specifically, during the model pre-training phase, we first apply the pseudo-labeled data examples to continuously pre-train UNITE. Notably, to reduce the gap between pre-training and fine-tuning, we use data cropping and a ranking-based score normalization strategy. During the fine-tuning phase, we use both Direct Assessment (DA) and Multidimensional Quality Metrics (MQM) data from past years' WMT competitions. Specially, we collect the results from models with different pre-trained language model backbones, and use different ensembling strategies for involved translation directions.