LLM Prompt Evaluation and Response Annotation
Scale AITextPrompt Response Writing SFT
Participated in a large-scale LLM training project focused on enhancing model accuracy, response relevance, and safety. Tasks included writing and evaluating prompts and responses for supervised fine-tuning (SFT), classifying entities, and rating model outputs for correctness, coherence, and tone. Involved in multi-layer review cycles to ensure high annotation consistency. Adhered to strict quality guidelines, including detailed rubric-based evaluation and linguistic accuracy checks across various user intent categories. The project contributed to improving the model's performance in question answering, conversational flow, and content moderation.
Participated in a large-scale LLM training project focused on enhancing model accuracy, response relevance, and safety. Tasks included writing and evaluating prompts and responses for supervised fine-tuning (SFT), classifying entities, and rating model outputs for correctness, coherence, and tone. Involved in multi-layer review cycles to ensure high annotation consistency. Adhered to strict quality guidelines, including detailed rubric-based evaluation and linguistic accuracy checks across various user intent categories. The project contributed to improving the model's performance in question answering, conversational flow, and content moderation.
2024 - 2024