Multilingual Few-Shot Learning via Language Model Retrieval
This work addresses the challenge of reliable few-shot learning for multilingual natural language understanding, though it is incremental as it builds on existing retrieval and in-context learning approaches.
The paper tackles the problem of high variability in few-shot learning performance due to example selection by proposing a retrieval-based method that uses semantically similar samples as context, which consistently outperforms random sampling in multilingual and cross-lingual tasks on non-English languages.
Transformer-based language models have achieved remarkable success in few-shot in-context learning and drawn a lot of research interest. However, these models' performance greatly depends on the choice of the example prompts and also has high variability depending on how samples are chosen. In this paper, we conduct a comprehensive study of retrieving semantically similar few-shot samples and using them as the context, as it helps the model decide the correct label without any gradient update in the multilingual and cross-lingual settings. We evaluate the proposed method on five natural language understanding datasets related to intent detection, question classification, sentiment analysis, and topic classification. The proposed method consistently outperforms random sampling in monolingual and cross-lingual tasks in non-English languages.