Picture for Marcin Junczys-Dowmunt

Marcin Junczys-Dowmunt

On-the-Fly Fusion of Large Language Models and Machine Translation

Nov 14, 2023
Viaarxiv icon

SOTASTREAM: A Streaming Approach to Machine Translation Training

Add code
Aug 14, 2023
Viaarxiv icon

Escaping the sentence-level paradigm in machine translation

Add code
Apr 25, 2023
Figure 1 for Escaping the sentence-level paradigm in machine translation
Figure 2 for Escaping the sentence-level paradigm in machine translation
Figure 3 for Escaping the sentence-level paradigm in machine translation
Figure 4 for Escaping the sentence-level paradigm in machine translation
Viaarxiv icon

The JHU-Microsoft Submission for WMT21 Quality Estimation Shared Task

Add code
Sep 17, 2021
Figure 1 for The JHU-Microsoft Submission for WMT21 Quality Estimation Shared Task
Figure 2 for The JHU-Microsoft Submission for WMT21 Quality Estimation Shared Task
Figure 3 for The JHU-Microsoft Submission for WMT21 Quality Estimation Shared Task
Figure 4 for The JHU-Microsoft Submission for WMT21 Quality Estimation Shared Task
Viaarxiv icon

Levenshtein Training for Word-level Quality Estimation

Sep 15, 2021
Figure 1 for Levenshtein Training for Word-level Quality Estimation
Figure 2 for Levenshtein Training for Word-level Quality Estimation
Figure 3 for Levenshtein Training for Word-level Quality Estimation
Figure 4 for Levenshtein Training for Word-level Quality Estimation
Viaarxiv icon

To Ship or Not to Ship: An Extensive Evaluation of Automatic Metrics for Machine Translation

Add code
Jul 22, 2021
Figure 1 for To Ship or Not to Ship: An Extensive Evaluation of Automatic Metrics for Machine Translation
Figure 2 for To Ship or Not to Ship: An Extensive Evaluation of Automatic Metrics for Machine Translation
Figure 3 for To Ship or Not to Ship: An Extensive Evaluation of Automatic Metrics for Machine Translation
Figure 4 for To Ship or Not to Ship: An Extensive Evaluation of Automatic Metrics for Machine Translation
Viaarxiv icon

On User Interfaces for Large-Scale Document-Level Human Evaluation of Machine Translation Outputs

Apr 21, 2021
Figure 1 for On User Interfaces for Large-Scale Document-Level Human Evaluation of Machine Translation Outputs
Figure 2 for On User Interfaces for Large-Scale Document-Level Human Evaluation of Machine Translation Outputs
Figure 3 for On User Interfaces for Large-Scale Document-Level Human Evaluation of Machine Translation Outputs
Figure 4 for On User Interfaces for Large-Scale Document-Level Human Evaluation of Machine Translation Outputs
Viaarxiv icon

The Curious Case of Hallucinations in Neural Machine Translation

Add code
Apr 14, 2021
Figure 1 for The Curious Case of Hallucinations in Neural Machine Translation
Figure 2 for The Curious Case of Hallucinations in Neural Machine Translation
Figure 3 for The Curious Case of Hallucinations in Neural Machine Translation
Figure 4 for The Curious Case of Hallucinations in Neural Machine Translation
Viaarxiv icon

Microsoft Translator at WMT 2019: Towards Large-Scale Document-Level Neural Machine Translation

Jul 14, 2019
Figure 1 for Microsoft Translator at WMT 2019: Towards Large-Scale Document-Level Neural Machine Translation
Figure 2 for Microsoft Translator at WMT 2019: Towards Large-Scale Document-Level Neural Machine Translation
Figure 3 for Microsoft Translator at WMT 2019: Towards Large-Scale Document-Level Neural Machine Translation
Figure 4 for Microsoft Translator at WMT 2019: Towards Large-Scale Document-Level Neural Machine Translation
Viaarxiv icon

Accelerating Asynchronous Stochastic Gradient Descent for Neural Machine Translation

Sep 14, 2018
Figure 1 for Accelerating Asynchronous Stochastic Gradient Descent for Neural Machine Translation
Figure 2 for Accelerating Asynchronous Stochastic Gradient Descent for Neural Machine Translation
Figure 3 for Accelerating Asynchronous Stochastic Gradient Descent for Neural Machine Translation
Figure 4 for Accelerating Asynchronous Stochastic Gradient Descent for Neural Machine Translation
Viaarxiv icon