HalOmi: A Manually Annotated Benchmark for Multilingual Hallucination and Omission Detection in Machine Translation.
David DaleElena VoitaJanice LamPrangthip HansantiChristophe RopersElahe KalbassiCynthia GaoLoïc BarraultMarta R. Costa-jussàPublished in: CoRR (2023)
Keyphrases
- machine translation
- manually annotated
- cross lingual
- language independent
- language resources
- cross language information retrieval
- relation extraction
- information extraction
- multilingual documents
- chinese english
- ground truth
- machine translation system
- parallel corpus
- target language
- natural language processing
- natural language
- word sense disambiguation
- word alignment
- cross language
- statistical machine translation
- translation model
- automatic extraction
- text classification
- machine learning
- source language
- parallel corpora
- digital libraries