FRMT: A Benchmark for Few-Shot Region-Aware Machine Translation.
Parker RileyTimothy DozatJan A. BothaXavier GarciaDan GarretteJason RiesaOrhan FiratNoah ConstantPublished in: Trans. Assoc. Comput. Linguistics (2023)
Keyphrases
- machine translation
- natural language processing
- information extraction
- statistical machine translation
- language independent
- cross lingual
- language processing
- cross language information retrieval
- natural language generation
- target language
- natural language
- chinese english
- word sense disambiguation
- word alignment
- video data
- video sequences
- visual features
- parallel corpora
- multilingual documents
- query translation
- machine translation system
- brazilian portuguese
- video shots
- key frames
- co occurrence
- finite state transducers
- statistical translation models