Papers by Spencer Rarrick
GATE X-E : A Challenge Set for Gender-Fair Translations from Weakly-Gendered Languages (2024.findings-acl)
Copied to clipboard
| Challenge: | despite improvements in NMT quality and adoption, perpetuation of gender bias remains a concern. |
| Approach: | They propose a translation gender rewriting solution built with GPT-4 and use it to evaluate it. |
| Outcome: | The proposed translation gender rewriting solution is open source and can be used in English. |
Combining Translation Memory with Neural Machine Translation (D19-52)
Copied to clipboard
| Challenge: | Existing systems that combine translation memory and statistical machine translation (MT) models are able to translate less familiar phrases and sentences without sacrificing quality. |
| Approach: | They propose to combine translation memory and Neural Machine Translation (NMT) models to select final translation outputs when similarity score of a test source sentence exceeds the predefined threshold. |
| Outcome: | The proposed system significantly improves translation performance on the Timely Disclosure corpus, as compared to a standalone NMT system. |