VOYAGER: A Training Free Approach for Generating Diverse Datasets using LLMs
This addresses the issue of limited dataset diversity for researchers and practitioners using LLMs for synthetic data generation, though it is an incremental improvement over existing methods.
The paper tackles the problem of low diversity in synthetic datasets generated by large language models by proposing Voyager, an iterative, training-free method that uses determinantal point processes to optimize diversity, achieving a 1.5-3x improvement over baseline approaches.
Large language models (LLMs) are increasingly being used to generate synthetic datasets for the evaluation and training of downstream models. However, prior work has noted that such generated data lacks diversity. In this paper, we propose Voyager, a novel principled approach to generate diverse datasets. Our approach is iterative and directly optimizes a mathematical quantity that optimizes the diversity of the dataset using the machinery of determinantal point processes. Furthermore, our approach is training-free, applicable to closed-source models, and scalable. In addition to providing theoretical justification for the working of our method, we also demonstrate through comprehensive experiments that Voyager significantly outperforms popular baseline approaches by providing a 1.5-3x improvement in diversity.