Investigating Contextual Influence in Document-Level Translation
Current state-of-the-art neural machine translation (NMT) architectures usually do not take document-level context into account. However, the document-level context of a source sentence to be translated could encode valuable information to guide the MT model to generate a better translation. In rece...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-05-01
|
| Series: | Information |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2078-2489/13/5/249 |
| _version_ | 1827668512324190208 |
|---|---|
| author | Prashanth Nayak Rejwanul Haque John D. Kelleher Andy Way |
| author_facet | Prashanth Nayak Rejwanul Haque John D. Kelleher Andy Way |
| author_sort | Prashanth Nayak |
| collection | DOAJ |
| description | Current state-of-the-art neural machine translation (NMT) architectures usually do not take document-level context into account. However, the document-level context of a source sentence to be translated could encode valuable information to guide the MT model to generate a better translation. In recent times, MT researchers have turned their focus to this line of MT research. As an example, hierarchical attention network (HAN) models use document-level context for translation prediction. In this work, we studied translations produced by the HAN-based MT systems. We examined how contextual information improves translation in document-level NMT. More specifically, we investigated why context-aware models such as HAN perform better than vanilla baseline NMT systems that do not take context into account. We considered Hindi-to-English, Spanish-to-English and Chinese-to-English for our investigation. We experimented with the formation of conditional context (i.e., neighbouring sentences) of the source sentences to be translated in HAN to predict their target translations. Interestingly, we observed that the quality of the target translations of specific source sentences highly relates to the context in which the source sentences appear. Based on their sensitivity to context, we classify our test set sentences into three categories, i.e., <i>context-sensitive</i>, <i>context-insensitive</i> and <i>normal</i>. We believe that this categorization may change the way in which context is utilized in document-level translation. |
| first_indexed | 2024-03-10T03:41:51Z |
| format | Article |
| id | doaj.art-c7c9cf1907734622b2d138724e864626 |
| institution | Directory Open Access Journal |
| issn | 2078-2489 |
| language | English |
| last_indexed | 2024-03-10T03:41:51Z |
| publishDate | 2022-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Information |
| spelling | doaj.art-c7c9cf1907734622b2d138724e8646262023-11-23T11:30:18ZengMDPI AGInformation2078-24892022-05-0113524910.3390/info13050249Investigating Contextual Influence in Document-Level TranslationPrashanth Nayak0Rejwanul Haque1John D. Kelleher2Andy Way3School of Computing, Dublin City University, D09 Y074 Dublin, IrelandSchool of Computing, National College of Ireland, D01 K6W2 Dublin, IrelandSchool of Computer Science, Technological University Dublin, D02 HW71 Dublin, IrelandSchool of Computing, Dublin City University, D09 Y074 Dublin, IrelandCurrent state-of-the-art neural machine translation (NMT) architectures usually do not take document-level context into account. However, the document-level context of a source sentence to be translated could encode valuable information to guide the MT model to generate a better translation. In recent times, MT researchers have turned their focus to this line of MT research. As an example, hierarchical attention network (HAN) models use document-level context for translation prediction. In this work, we studied translations produced by the HAN-based MT systems. We examined how contextual information improves translation in document-level NMT. More specifically, we investigated why context-aware models such as HAN perform better than vanilla baseline NMT systems that do not take context into account. We considered Hindi-to-English, Spanish-to-English and Chinese-to-English for our investigation. We experimented with the formation of conditional context (i.e., neighbouring sentences) of the source sentences to be translated in HAN to predict their target translations. Interestingly, we observed that the quality of the target translations of specific source sentences highly relates to the context in which the source sentences appear. Based on their sensitivity to context, we classify our test set sentences into three categories, i.e., <i>context-sensitive</i>, <i>context-insensitive</i> and <i>normal</i>. We believe that this categorization may change the way in which context is utilized in document-level translation.https://www.mdpi.com/2078-2489/13/5/249machine translationneural machine translationcontext-aware translationdocument translation |
| spellingShingle | Prashanth Nayak Rejwanul Haque John D. Kelleher Andy Way Investigating Contextual Influence in Document-Level Translation Information machine translation neural machine translation context-aware translation document translation |
| title | Investigating Contextual Influence in Document-Level Translation |
| title_full | Investigating Contextual Influence in Document-Level Translation |
| title_fullStr | Investigating Contextual Influence in Document-Level Translation |
| title_full_unstemmed | Investigating Contextual Influence in Document-Level Translation |
| title_short | Investigating Contextual Influence in Document-Level Translation |
| title_sort | investigating contextual influence in document level translation |
| topic | machine translation neural machine translation context-aware translation document translation |
| url | https://www.mdpi.com/2078-2489/13/5/249 |
| work_keys_str_mv | AT prashanthnayak investigatingcontextualinfluenceindocumentleveltranslation AT rejwanulhaque investigatingcontextualinfluenceindocumentleveltranslation AT johndkelleher investigatingcontextualinfluenceindocumentleveltranslation AT andyway investigatingcontextualinfluenceindocumentleveltranslation |