Joint translation and unit conversion for end-to-end localization
This addresses the challenge of mixed-language text processing for applications requiring localization, but it appears incremental as it builds on existing translation and conversion methods.
The paper tackled the problem of processing text containing both natural language and formal languages like mathematical expressions, using unit conversions as an example, and proposed a data augmentation technique that enables models to learn translation and conversion tasks and switch between them for end-to-end localization.
A variety of natural language tasks require processing of textual data which contains a mix of natural language and formal languages such as mathematical expressions. In this paper, we take unit conversions as an example and propose a data augmentation technique which leads to models learning both translation and conversion tasks as well as how to adequately switch between them for end-to-end localization.