Distilling Translations with Visual Awareness
- Submitting institution
-
The University of Sheffield
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 7853
- Type
- E - Conference contribution
- DOI
-
10.18653/v1/p19-1653
- Title of conference / published proceedings
- Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics
- First page
- 6525
- Volume
- -
- Issue
- -
- ISSN
- -
- Open access status
- -
- Month of publication
- July
- Year of publication
- 2019
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
2
- Research group(s)
-
D - Natural Language Processing
- Citation count
- 3
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- We propose a novel approach to multimodal machine translation that is inspired by how humans translate: they use the source text to generate a draft translation and then refine it based on additional context – here, visual information. This approach, accepted in the A* Ranked ACL conference, led to the state-of-the-art performance in benchmarking datasets and a research gift ($ 100,000) from the US Air Force Research Lab.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -