Meta-Learning for Effective Multi-task and Multilingual Modelling
This work addresses the challenge of multi-task and multilingual modeling for NLP applications, offering a novel integration that enhances cross-lingual and cross-task learning.
The paper tackled the problem of learning shared representations across tasks and languages in NLP by proposing a meta-learning approach, resulting in clear performance improvements on the XTREME benchmark and effective zero-shot evaluations on unseen languages.
Natural language processing (NLP) tasks (e.g. question-answering in English) benefit from knowledge of other tasks (e.g. named entity recognition in English) and knowledge of other languages (e.g. question-answering in Spanish). Such shared representations are typically learned in isolation, either across tasks or across languages. In this work, we propose a meta-learning approach to learn the interactions between both tasks and languages. We also investigate the role of different sampling strategies used during meta-learning. We present experiments on five different tasks and six different languages from the XTREME multilingual benchmark dataset. Our meta-learned model clearly improves in performance compared to competitive baseline models that also include multi-task baselines. We also present zero-shot evaluations on unseen target languages to demonstrate the utility of our proposed model.