Alert button
Picture for Danielle Saunders

Danielle Saunders

Alert button

Gender, names and other mysteries: Towards the ambiguous for gender-inclusive translation

Jun 07, 2023
Danielle Saunders, Katrina Olsen

Figure 1 for Gender, names and other mysteries: Towards the ambiguous for gender-inclusive translation
Figure 2 for Gender, names and other mysteries: Towards the ambiguous for gender-inclusive translation
Figure 3 for Gender, names and other mysteries: Towards the ambiguous for gender-inclusive translation
Figure 4 for Gender, names and other mysteries: Towards the ambiguous for gender-inclusive translation
Viaarxiv icon

First the worst: Finding better gender translations during beam search

Apr 15, 2021
Danielle Saunders, Rosie Sallis, Bill Byrne

Figure 1 for First the worst: Finding better gender translations during beam search
Figure 2 for First the worst: Finding better gender translations during beam search
Figure 3 for First the worst: Finding better gender translations during beam search
Figure 4 for First the worst: Finding better gender translations during beam search
Viaarxiv icon

Domain Adaptation and Multi-Domain Adaptation for Neural Machine Translation: A Survey

Apr 14, 2021
Danielle Saunders

Figure 1 for Domain Adaptation and Multi-Domain Adaptation for Neural Machine Translation: A Survey
Figure 2 for Domain Adaptation and Multi-Domain Adaptation for Neural Machine Translation: A Survey
Figure 3 for Domain Adaptation and Multi-Domain Adaptation for Neural Machine Translation: A Survey
Figure 4 for Domain Adaptation and Multi-Domain Adaptation for Neural Machine Translation: A Survey
Viaarxiv icon

Inference-only sub-character decomposition improves translation of unseen logographic characters

Nov 12, 2020
Danielle Saunders, Weston Feely, Bill Byrne

Figure 1 for Inference-only sub-character decomposition improves translation of unseen logographic characters
Figure 2 for Inference-only sub-character decomposition improves translation of unseen logographic characters
Figure 3 for Inference-only sub-character decomposition improves translation of unseen logographic characters
Figure 4 for Inference-only sub-character decomposition improves translation of unseen logographic characters
Viaarxiv icon

Addressing Exposure Bias With Document Minimum Risk Training: Cambridge at the WMT20 Biomedical Translation Task

Oct 11, 2020
Danielle Saunders, Bill Byrne

Figure 1 for Addressing Exposure Bias With Document Minimum Risk Training: Cambridge at the WMT20 Biomedical Translation Task
Figure 2 for Addressing Exposure Bias With Document Minimum Risk Training: Cambridge at the WMT20 Biomedical Translation Task
Figure 3 for Addressing Exposure Bias With Document Minimum Risk Training: Cambridge at the WMT20 Biomedical Translation Task
Figure 4 for Addressing Exposure Bias With Document Minimum Risk Training: Cambridge at the WMT20 Biomedical Translation Task
Viaarxiv icon

Neural Machine Translation Doesn't Translate Gender Coreference Right Unless You Make It

Oct 11, 2020
Danielle Saunders, Rosie Sallis, Bill Byrne

Figure 1 for Neural Machine Translation Doesn't Translate Gender Coreference Right Unless You Make It
Figure 2 for Neural Machine Translation Doesn't Translate Gender Coreference Right Unless You Make It
Figure 3 for Neural Machine Translation Doesn't Translate Gender Coreference Right Unless You Make It
Figure 4 for Neural Machine Translation Doesn't Translate Gender Coreference Right Unless You Make It
Viaarxiv icon

Using Context in Neural Machine Translation Training Objectives

May 04, 2020
Danielle Saunders, Felix Stahlberg, Bill Byrne

Figure 1 for Using Context in Neural Machine Translation Training Objectives
Figure 2 for Using Context in Neural Machine Translation Training Objectives
Figure 3 for Using Context in Neural Machine Translation Training Objectives
Figure 4 for Using Context in Neural Machine Translation Training Objectives
Viaarxiv icon

Reducing Gender Bias in Neural Machine Translation as a Domain Adaptation Problem

Apr 21, 2020
Danielle Saunders, Bill Byrne

Figure 1 for Reducing Gender Bias in Neural Machine Translation as a Domain Adaptation Problem
Figure 2 for Reducing Gender Bias in Neural Machine Translation as a Domain Adaptation Problem
Figure 3 for Reducing Gender Bias in Neural Machine Translation as a Domain Adaptation Problem
Figure 4 for Reducing Gender Bias in Neural Machine Translation as a Domain Adaptation Problem
Viaarxiv icon

UCAM Biomedical translation at WMT19: Transfer learning multi-domain ensembles

Jun 13, 2019
Danielle Saunders, Felix Stahlberg, Bill Byrne

Figure 1 for UCAM Biomedical translation at WMT19: Transfer learning multi-domain ensembles
Figure 2 for UCAM Biomedical translation at WMT19: Transfer learning multi-domain ensembles
Figure 3 for UCAM Biomedical translation at WMT19: Transfer learning multi-domain ensembles
Figure 4 for UCAM Biomedical translation at WMT19: Transfer learning multi-domain ensembles
Viaarxiv icon