HalOmi: A Manually Annotated Benchmark for Multilingual Hallucination and Omission Detection in Machine Translation
This provides a resource for improving error detection in machine translation across diverse languages, addressing a critical issue for users and developers, though it is incremental as it builds on existing annotation efforts.
The authors tackled the scarcity of annotated data for hallucinations and omissions in machine translation by releasing a manually annotated benchmark covering 18 translation directions, and they established new baselines by showing that previous detection methods do not generalize well across languages.
Hallucinations in machine translation are translations that contain information completely unrelated to the input. Omissions are translations that do not include some of the input information. While both cases tend to be catastrophic errors undermining user trust, annotated data with these types of pathologies is extremely scarce and is limited to a few high-resource languages. In this work, we release an annotated dataset for the hallucination and omission phenomena covering 18 translation directions with varying resource levels and scripts. Our annotation covers different levels of partial and full hallucinations as well as omissions both at the sentence and at the word level. Additionally, we revisit previous methods for hallucination and omission detection, show that conclusions made based on a single language pair largely do not hold for a large-scale evaluation, and establish new solid baselines.