Are pre-trained text representations useful for multilingual and multi-dimensional language proficiency modeling?
This work addresses the incremental improvement of language proficiency modeling for non-native learners by extending it to multiple languages and dimensions, though it highlights limitations in feature consistency.
The paper tackled the problem of modeling language proficiency for non-native learners by exploring pre-trained and fine-tuned multilingual embeddings for multi-dimensional classification across German, Italian, and Czech, finding that fine-tuned embeddings are useful but no features consistently perform best across all seven proficiency dimensions.
Development of language proficiency models for non-native learners has been an active area of interest in NLP research for the past few years. Although language proficiency is multidimensional in nature, existing research typically considers a single "overall proficiency" while building models. Further, existing approaches also considers only one language at a time. This paper describes our experiments and observations about the role of pre-trained and fine-tuned multilingual embeddings in performing multi-dimensional, multilingual language proficiency classification. We report experiments with three languages -- German, Italian, and Czech -- and model seven dimensions of proficiency ranging from vocabulary control to sociolinguistic appropriateness. Our results indicate that while fine-tuned embeddings are useful for multilingual proficiency modeling, none of the features achieve consistently best performance for all dimensions of language proficiency. All code, data and related supplementary material can be found at: https://github.com/nishkalavallabhi/MultidimCEFRScoring.