RLHF Evaluator | Finance Data Trainer | ESG / Carbon Accounting Reviewer | Chinese-English AI Response Evaluator
Text
RLHFAs an RLHF Evaluator and Bilingual AI Response Reviewer, I performed data annotation by ranking, scoring, and critiquing AI-generated responses in both English and Chinese. I evaluated model outputs for financial and ESG accuracy, logical consistency, and adherence to compliance and disclosure standards. My workflow included detecting hallucinations, flagging unsafe or misleading outputs, and improving responses to better align with user needs. • Conducted rubric-based evaluation and ranking of large language model outputs. • Provided bilingual (English/Chinese) review for cultural and linguistic accuracy. • Fact-checked financial, accounting, and ESG-related AI answers. • Used prompting and specialized workflows for quality assurance and training improvement.