ViDeBERTa: A powerful pre-trained language model for Vietnamese.
Cong Dao TranNhut Huy PhamAnh NguyenTruong Son HyTu VuPublished in: EACL (Findings) (2023)
Keyphrases
- language model
- pre trained
- language modeling
- document representation
- n gram
- document retrieval
- information retrieval
- probabilistic model
- training data
- context sensitive
- query expansion
- retrieval model
- training examples
- test collection
- speech recognition
- control signals
- smoothing methods
- vector space model
- translation model
- word clouds
- ad hoc information retrieval
- relevance feedback
- relevance model
- cross lingual
- small number
- feature space
- text mining
- image classification