New framework optimizes LLM-augmented surveys using human respondent allocation

A new research paper published on arXiv introduces a framework for optimizing survey design when using large language models (LLMs) for response generation. The framework addresses the challenge of LLM accuracy variability by characterizing a question-specific "rectification difficulty" and proposing an optimal allocation rule for human respondents. This approach aims to direct human labeling efforts to tasks where LLMs are least reliable, thereby improving overall survey efficiency and accuracy without requiring pilot human data. AI

IMPACT This research could lead to more efficient and accurate survey designs by optimizing the use of LLMs and human annotators.

RANK_REASON The cluster contains a research paper detailing a new framework for LLM-augmented surveys. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• alphaXiv
• arXiv
• DagsHub
• Gotit.pub
• Hugging Face
• large-language models
• Prediction-powered inference
• ScienceCast
• Zikun Ye

• paper
• product

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →