Do Multilingual Language Models Think Better in English?

Laburpena

Translate-test is a popular technique to improve the performance of multilingual language models. This approach works by translating the input into English using an external machine translation system, and running inference over the translated input. However, these improvements can be attributed to the use of a separate translation system, which is typically trained on large amounts of parallel data not seen by the language model. In this work, we introduce a new approach called self-translate, which overcomes the need of an external translation system by leveraging the few-shot translation capabilities of multilingual language models. Experiments over 5 tasks show that self-translate consistently outperforms direct inference, demonstrating that language models are unable to leverage their full multilingual potential when prompted in non-English languages. Our code is available at https://github.com/juletx/self-translate.

Argitalpena
NAACL 2024
Julen Etxaniz
Julen Etxaniz
Hizkuntzaren Azterketa eta Prozesamendua Doktoregoko ikaslea

Hizkuntzaren Azterketa eta Prozesamendua Doktoregoko ikaslea HiTZ Zentroko IXA taldean (UPV/EHU). Hizkuntza ereduak baliabidea urriko hizkuntzetako hobetzeko lanean. Informatika Ingeniaritzan graduatua Software Ingeniaritza espezialitatearekin. Hizkuntzaren Azterketa eta Prozesamendua Masterra.