Validating Large Language Model–Assisted Qualitative Research: Ensuring Methodological Integrity in Engineering Education Research

Engineering notes

Key topics: autonomous driving, large language model. See the paper for implementation details and experimental results.

Chinese explanation / 中文解读

中文解读待补充：本站会优先为端到端自动驾驶、BEV感知、3D目标检测、轨迹预测、路径规划、LiDAR感知等高价值论文补充中文说明。

Original abstract

There are long-standing, accepted methods for validating analysis of textual data. However, there has always existed limitations on the scale of data to which human analysis can be applied. Qualitative researchers are now incorporating Large Language Models (LLMs) in their data analysis processes and beginning to standardize how LLMs can be used to reliably analyze larger sets of qualitative data, without losing the inherent reflexive nature necessary for high quality analysis. The goal of this paper is to contribute to the broader understanding of validation methods emerging for qualitative research that rely on LLMs. First, a review of the literature illustrates similarities in validation techniques LLM outputs. We then document the approach we took in our research, which explores the exposure of engineering tasks to applications of artificial intelligence. This provides a practical example of validation in a context that is different from using an LLM for qualitative coding.

Links and sources

• PDF from original source

Comments