Japanese LLM Response Evaluation and Prompt Annotation
Scale AITextPrompt Response Writing SFT
Worked on Japanese-language data labeling and evaluation tasks for LLM training.
Responsibilities included evaluating pairs of model-generated responses across multiple quality dimensions such as instruction adherence, localization accuracy, and factual correctness.
Additionally performed prompt annotation by creating text and spoken prompts on predefined topics, as well as producing accurate audio transcriptions following strict annotation guidelines.
All tasks were completed using client-provided annotation platforms and subjected to quality control standards.
Worked on Japanese-language data labeling and evaluation tasks for LLM training. Responsibilities included evaluating pairs of model-generated responses across multiple quality dimensions such as instruction adherence, localization accuracy, and factual correctness. Additionally performed prompt annotation by creating text and spoken prompts on predefined topics, as well as producing accurate audio transcriptions following strict annotation guidelines. All tasks were completed using client-provided annotation platforms and subjected to quality control standards.
2024