French AI Responses Evaluation and Rating
This project involved evaluating and rating 2 AI responses in French, as well as reviewing and correcting other freelancers' work to ensure guidelines were followed. Both AI models had to be rated on Localization, Truthfulness, Instruction Following, Harmlessness, and Writing Quality. Both responses then had to be compared to each other with a comparative rating to determine if one model was better. Reviewing others' work involved verifying if the ratings given to both models were accurate and made sense according the justification given by the annotator.