GreenPLM: Cross-lingual pre-trained language models conversion with (almost) no cost.
Qingcheng ZengLucas GarayPeilin ZhouDading ChongYining HuaJiageng WuYikang PanHan ZhouJie YangPublished in: CoRR (2022)
Keyphrases
- language modeling
- cross lingual
- language model
- pre trained
- translation model
- pseudo feedback
- n gram
- retrieval model
- probabilistic model
- information retrieval
- language independent
- document retrieval
- cross lingual information retrieval
- speech recognition
- cross language
- query expansion
- test collection
- training data
- language modeling framework
- word segmentation
- cross language retrieval
- query terms
- text classification
- relevance model
- parallel corpora
- out of vocabulary
- learning process
- pseudo relevance feedback
- data sets
- machine translation