Skip to content
conftrace_
2023 EMNLP EMNLP 2023

PRHLT’s Submission to WLAC 2023

Abstract

AbstractThis paper describes our submission to the Word-Level AutoCompletion shared task of WMT23. We participated in the English–German and German–English categories. We extended our last year segment-based interactive machine translation approach to address its weakness when no context is available. Additionally, we fine-tune the pre-trained mT5 large language model to be used for autocompletion.

🌉 Interdisciplinary Bridge - Artificial Intelligence and Natural Language Processing

Authors

Angel Navarro, Miguel Domingo, Francisco Casacuberta

Topics

Deep Learning > Techniques > Pretraining Natural Language Processing > Generation > Text Generation Natural Language Processing > Applications > Machine Translation Machine Learning > Learning Types > Transfer Learning

Keywords

language model pre-trained language model interactive machine translation word-level auto-completion
Download PDF

More from EMNLP 2023

All EMNLP 2023 papers →
MMT’s Submission for the WMT 2023 Quality Estimation Shared Task
IOL Research’s Submission for WMT 2023 Quality Estimation Shared Task
SJTU-MTLAB’s Submission to the WMT23 Word-Level Auto Completion Task
KnowComp Submission for WMT23 Word-Level AutoCompletion Task
Terminology-Aware Translation with Constrained Decoding and Large Language Model Prompting
Lingua Custodia’s Participation at the WMT 2023 Terminology Shared Task