November 24, 2024

How well do large language models perform in green building assessment? An evaluation analysis from a case data set

View Full Paper

Ask AI

Bookmark

Authors

Discussion

Analysis reveals LLMs improve efficiency in green building assessments, suggesting ways to enhance performance.

LLMs excel in green building assessments, particularly with short text inputs, as seen with GPT-4.
Performance metrics indicate GPT-4's highest effectiveness, especially when combined with prompt engineering techniques.
Observational analysis utilizing a dataset of 1200 real-world cases, assessing both text input types and LLMs.
Findings suggest that while LLMs handle qualitative criteria well, they struggle with quantitative assessments requiring complex calculations.

Qiufeng He (2024) studied this question.

View Full Paper

Ask AI

Bookmark