To Translate or Not to Translate: A Systematic Investigation of Translation-Based Cross-Lingual Transfer to Low-Resource Languages
This work addresses the problem of improving cross-lingual transfer for low-resource languages, offering incremental but practical advancements in multilingual NLP.
The paper systematically investigates translation-based cross-lingual transfer to low-resource languages, finding that these approaches dramatically outperform zero-shot methods, with round-trip translation combined with test translation being most effective, achieving significant empirical gains.
Perfect machine translation (MT) would render cross-lingual transfer (XLT) by means of multilingual language models (mLMs) superfluous. Given, on the one hand, the large body of work on improving XLT with mLMs and, on the other hand, recent advances in massively multilingual MT, in this work, we systematically evaluate existing and propose new translation-based XLT approaches for transfer to low-resource languages. We show that all translation-based approaches dramatically outperform zero-shot XLT with mLMs -- with the combination of round-trip translation of the source-language training data and the translation of the target-language test instances at inference -- being generally the most effective. We next show that one can obtain further empirical gains by adding reliable translations to other high-resource languages to the training data. Moreover, we propose an effective translation-based XLT strategy even for languages not supported by the MT system. Finally, we show that model selection for XLT based on target-language validation data obtained with MT outperforms model selection based on the source-language data. We believe our findings warrant a broader inclusion of more robust translation-based baselines in XLT research.