EMNLP Findings2023
Investigating the Effectiveness of Multiple Expert Models Collaboration
Abstract
This paper aims to investigate the effectiveness of several machine translation (MT) models and aggregation methods in a multi-domain setting under fair conditions and explore a direction for tackling multi-domain MT. We mainly compare the performance of the single model approach by jointly training all domains and the multi-expert models approach with a particular aggregation strategy. We conduct experiments on multiple domain datasets and demonstrate that a combination of smaller domain expert models can outperform a larger model trained for all domain data.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Lineage
Builds on3
- BLEURT: Learning Robust Metrics for Text GenerationThibault Sellam, Dipanjan Das, Ankur P. ParikhACL 2020 · 40 citations
- Multi-Domain Neural Machine Translation with Word-Level Adaptive Layer-wise Domain MixingHaoming Jiang, Chen Liang, Chong Wang, Tuo ZhaoACL 2020 · 26 citations
- COMET: A Neural Framework for MT EvaluationRicardo Rei, Craig Stewart, Ana C. Farinha, Alon LavieEMNLP 2020 · 6 citations
Related papers
- Distilling Multiple Domains for Neural Machine TranslationAnna Currey, Prashant Mathur, Georgiana DinuEMNLP 2020 · 19 citations
- Go From the General to the Particular: Multi-Domain Translation with Domain Transformation NetworksYong Wang, Longyue Wang, Shuming Shi, Victor O. K. Li et al.AAAI 2020 · 30 citations
- Uncertainty-Aware Balancing for Multilingual and Multi-Domain Neural Machine Translation TrainingMinghao Wu, Yitong Li, Meng Zhang, Liangyou Li et al.EMNLP 2021 · 10 citations
- Learning a Multi-Domain Curriculum for Neural Machine TranslationWei Wang, Ye Tian, Jiquan Ngiam, Yinfei Yang et al.ACL 2020 · 32 citations
- Factorized Transformer for Multi-Domain Neural Machine TranslationYongchao Deng, Hongfei Yu, Heng Yu, Xiangyu Duan et al.EMNLP Findings 2020 · 5 citations
