Skip to content
conftrace_
2022 EMNLP EMNLP 2022

Partial Could Be Better than Whole. HW-TSC 2022 Submission for the Metrics Shared Task

Abstract

AbstractIn this paper, we present the contribution of HW-TSC to WMT 2022 Metrics Shared Task. We propose one reference-based metric, HWTSC-EE-BERTScore*, and four referencefree metrics including HWTSC-Teacher-Sim, HWTSC-TLM, KG-BERTScore and CROSSQE. Among these metrics, HWTSC-Teacher-Sim and CROSS-QE are supervised, whereas HWTSC-EE-BERTScore*, HWTSC-TLM and KG-BERTScore are unsupervised. We use these metrics in the segment-level and systemlevel tracks. Overall, our systems achieve strong results for all language pairs on previous test sets and a new state-of-the-art in many sys-level case sets.

Authors

Yilun Liu, Xiaosong Qiao, Zhanglin Wu, Su Chang, Min Zhang, Yanqing Zhao, Song Peng, Shimin Tao, Hao Yang, Ying Qin, Jiaxin Guo, Minghan Wang, Yinglu Li, Peng Li, Xiaofeng Zhao

Topics

Machine Learning > Core Methods > Representation Learning Machine Learning > Optimization & Theory > Optimization

Keywords

neural machine translation bert score teacher-student learning cross-lingual evaluation reference metric
Download PDF

More from EMNLP 2022

All EMNLP 2022 papers →
Linguistically Motivated Evaluation of Machine Translation Metrics Based on a Challenge Set
Exploring Robustness of Machine Translation Metrics: A Study of Twenty-Two Automatic Metrics in the WMT22 Metric Task
MS-COMET: More and Better Human Judgements Improve Metric Performance
Unsupervised Embedding-based Metric for MT Evaluation with Improved Human Correlation
REUSE: REference-free UnSupervised Quality Estimation Metric
MaTESe: Machine Translation Evaluation as a Sequence Tagging Problem