Don't Stop Fine-Tuning: On Training Regimes for Few-Shot Cross-Lingual Transfer with Multilingual Language Models.
Fabian David SchmidtIvan VulicGoran GlavasPublished in: EMNLP (2022)
Keyphrases
- cross lingual
- language modeling
- language model
- fine tuning
- cross lingual information retrieval
- cross language
- translation model
- document retrieval
- retrieval model
- n gram
- pseudo feedback
- language independent
- information retrieval
- probabilistic model
- query expansion
- transfer learning
- parallel corpus
- training set
- vector space model
- test collection
- machine learning
- smoothing methods
- query terms
- text classification
- out of vocabulary
- word segmentation
- relevance model
- pseudo relevance feedback
- supervised learning
- training data
- statistical machine translation
- visual features
- multimedia
- cross language retrieval
- machine translation
- language modeling framework