Task description: 

(1) Writing Task: You will be shown a text (i.e. starting of a paragraph/ context). You need to complete the paragraph with 50-100 words.

(2) Annotation Task: You will be shown texts in Fula (ADLaM script) corresponding to a writing assistant’s output based on a user’s transcription. Each text will have a user’s transcription (in one paragraph, which may or may not contain specific user requests) as grounding context. The output should organize the transcription into a well-structured document. Your job will be to judge each of the texts based on the following metrics:

1.Linguistic acceptability (does the text sound right?)

2.Quality of the output with respect to the feature (did the model do what it was instructed to do?)

3.Hallucinations (is there any invented content?)

4.Low-quality content (what is the general quality of the output?)

Work load: Up to 25 hours to complete all the sentences available

Schedule: Expected Project Duration: June 2025

Estimated time to complete the task: up to 25 hours (10-12h per week)

Rate: 20USD per hour

Requirements:

To be eligible to be considered for the Project you must meet the following criteria:

Please note that you are advised to only apply to one project at a time and to wait for confirmation regarding the success of your application before applying to another project. Applying to many projects at the same time might delay the revision of your application. Thank you very much for your understanding.

Raters who relocate to another country during their participation should always inform the production team of their move.

How to Apply:

Application: complete the application and our team will contact you with the next steps.

If you have any questions or would like more details about this project, please contact us at aisourcing@telusinternational.ai and please include the code ["127396 - Writing and Annotation Project - FULA (ADLaM)"] in the subject line.