Ethics and sustainability in machine translation. Research Seminar
![Thumbnail](/xmlui/bitstream/handle/10234/202908/Ethics%20and%20sustainability%20in%20machine%20translation.pdf.jpg?sequence=5&isAllowed=y)
Metadata
Show full item recordcomunitat-uji-handle:10234/192780
comunitat-uji-handle2:10234/192784
comunitat-uji-handle3:
comunitat-uji-handle4:
UJIMetadata
Title
Ethics and sustainability in machine translation. Research Seminar Date
2023-05-15Abstract
Neural machine translation (MT) can facilitate communication in a way that surpasses
previous MT paradigms, but there are also consequences of its use. As with the development of any technology, MT is not ethically ... [+]
Neural machine translation (MT) can facilitate communication in a way that surpasses
previous MT paradigms, but there are also consequences of its use. As with the development of any technology, MT is not ethically neutral, but rather reflects the
values of those behind its development. This is a particularly important consideration
when using technology that is predictive.
Conventional notions of validity and reliability are based on testable and consistent
results, yet these cannot apply to machine learning, and thus considerations regarding
reliability tend to be overlooked. Well-designed products are expected to be almost
invisible, an extension of the user, but again this is not possible
with unpredictable technologies such as MT (among other products based on large
language models). This presentation reflects on ethical issues around MT, beginning
with data gathering and reuse and looking at how MT fits with the values and codes of
the translator.
The computing power requirements of neural networks also behoves us to think about
issues of environmental as well as social sustainability related to the deployment of
MT. Rather than promoting an approach that involves following a set of instructions to
implement a technology unthinkingly, contemporary thinking in this area highlights the
importance of a conscious decision-making process when designing a data-driven MT
workflow. [-]
Subject
Type
info:eu-repo/semantics/conferenceObject; VideoPublisher version
https://mailchi.mp/0c29bbba47c0/machine-translationRights
http://rightsstatements.org/vocab/CNE/1.0/
info:eu-repo/semantics/openAccess
info:eu-repo/semantics/openAccess