AI Trainer and Evaluator for RLHF/Chain-of-Thought (CoT) in Life Sciences
TextRLHF
Applied scientific logic to Reinforcement Learning from Human Feedback (RLHF) tasks for validation of large language model responses. Focused on detection of biological hallucinations and factual validation of AI-generated protocols in regenerative medicine. Utilized Chain-of-Thought (CoT) validation and fact-checking expertise to review and rate AI-generated text output.
• Validated evidence chains and provided feedback on LLMs regarding biological and scientific queries.
• Ensured accuracy and consistency in AI-generated research protocols and model responses.
• Wrote rationales and engineered prompts for model evaluation and improvement.
• Employed Google Prompting Essentials and related tools to facilitate effective RLHF tasks.
Applied scientific logic to Reinforcement Learning from Human Feedback (RLHF) tasks for validation of large language model responses. Focused on detection of biological hallucinations and factual validation of AI-generated protocols in regenerative medicine. Utilized Chain-of-Thought (CoT) validation and fact-checking expertise to review and rate AI-generated text output. • Validated evidence chains and provided feedback on LLMs regarding biological and scientific queries. • Ensured accuracy and consistency in AI-generated research protocols and model responses. • Wrote rationales and engineered prompts for model evaluation and improvement. • Employed Google Prompting Essentials and related tools to facilitate effective RLHF tasks.
2014 - Present