The LLM Data Auditor: A Metric-oriented Survey on Quality and Trustworthiness in Evaluating Synthetic Data
This work addresses the critical problem of evaluating synthetic data quality for researchers and practitioners in AI, but it is incremental as it builds on existing generation methodologies by shifting focus to intrinsic evaluation metrics.
The paper tackles the challenge of ensuring high quality in LLM-generated synthetic data by proposing the LLM Data Auditor framework, which systematically categorizes intrinsic metrics for evaluating data quality and trustworthiness across six modalities, identifying substantial deficiencies in current evaluation practices and offering concrete recommendations for improvement.
Large Language Models (LLMs) have emerged as powerful tools for generating data across various modalities. By transforming data from a scarce resource into a controllable asset, LLMs mitigate the bottlenecks imposed by the acquisition costs of real-world data for model training, evaluation, and system iteration. However, ensuring the high quality of LLM-generated synthetic data remains a critical challenge. Existing research primarily focuses on generation methodologies, with limited direct attention to the quality of the resulting data. Furthermore, most studies are restricted to single modalities, lacking a unified perspective across different data types. To bridge this gap, we propose the \textbf{LLM Data Auditor framework}. In this framework, we first describe how LLMs are utilized to generate data across six distinct modalities. More importantly, we systematically categorize intrinsic metrics for evaluating synthetic data from two dimensions: quality and trustworthiness. This approach shifts the focus from extrinsic evaluation, which relies on downstream task performance, to the inherent properties of the data itself. Using this evaluation system, we analyze the experimental evaluations of representative generation methods for each modality and identify substantial deficiencies in current evaluation practices. Based on these findings, we offer concrete recommendations for the community to improve the evaluation of data generation. Finally, the framework outlines methodologies for the practical application of synthetic data across different modalities.