Researchers from UOL@IDEM have detailed their submission for the BEA 2026 shared task on L1-aware vocabulary difficulty prediction. Their approach models the task as a regression problem, training separate systems for Spanish, German, and Chinese. The system integrates multilingual contextual representations with engineered features, achieving RMSE scores of 1.132 for Spanish, 1.037 for German, and 0.891 for Chinese. AI
IMPACT This research contributes to better understanding and modeling of vocabulary difficulty across different languages, potentially aiding in language learning tools and educational content creation.
RANK_REASON The cluster contains a research paper detailing a submission to a shared task.
- BEA 2026
- BGE M3-Embedding: Multi-Lingual, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillation
- German
- LaBSE
- multilingual E5
- Spanish
- Standard Chinese
- UOL@IDEM
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →