Multilingual Machine Translation Evaluation Metrics Fine-tuned on Pseudo-Negative Examples for WMT 2021 Metrics Task.
Kosuke TakahashiYoichi IshibashiKatsuhito SudohSatoshi NakamuraPublished in: WMT@EMNLP (2021)
Keyphrases
- evaluation metrics
- negative examples
- fine tuned
- machine translation
- lexical cohesion
- cross lingual
- positive examples
- cross language information retrieval
- language resources
- chinese english
- precision and recall
- training data
- binary classification
- training examples
- machine translation system
- information extraction
- evaluation measures
- natural language processing
- background knowledge
- cross language
- query translation
- learning to rank
- domain specific
- target language
- digital libraries
- natural language
- parallel corpora
- multimedia
- information retrieval
- learning algorithm
- statistical machine translation
- training set