Our client is a trusted data partner and a leader in AI training data, providing high-quality, diverse datasets that power AI models. They specialize in creating human-generated data to train, fine-tune, and evaluate AI models across multiple domains.
They are currently looking for bilingual speakers with translation experience to join the team in evaluating AI translations.
Job Description
This project is dedicated to evaluating the accuracy and semantic quality of machine-translated sentences across multiple languages. This task involves evaluating how accurately the translated paragraph conveys the intended meaning of the original (source) paragraph. The goal is to determine the degree of semantic alignment between the two texts i.e., whether the ideas, information, and intent remain consistent after translation.
- Read both the source and translated paragraphs.
- Compare them sentence by sentence.
- Focus on semantic meaning only (not grammar, fluency, or tone).
- Judge whether each translated sentence expresses the same idea as the original.
- Assign a semantic equivalence score
- Add short comments for any mismatched or unclear meanings
Qualifications
- Native speaker in both the Mandarin Chinese and Amis languages is preferred but fluent speakers can also be considered as long as they can assess the accuracy of the languages and translations.
- With translation experience in the targeted languages.
Additional Information
Remote Opportunity!
Flexible hours!
