paper-with-me

Papers

Measuring Diversity in Synthetic Datasets

2025-02-12 · Yuchang Zhu, Huizhe Zhang, Bingzhe Wu, Jintang Li, Zibin Zheng, Peilin Zhao, Liang Chen, Yatao Bian

Large language models (LLMs) are widely adopted to generate synthetic datasets for various natural language processing (NLP) tasks, such as text classification and summarization. However, accurately measuring the diversity of these synthetic datasets-an aspect crucial for robust model performance-remains a significant challenge. In this paper, we introduce DCScore, a novel method for measuring synthetic dataset diversity from a classification perspective. Specifically, DCScore formulates diversity evaluation as a sample classification task, leveraging mutual relationships among samples. We further provide theoretical verification of the diversity-related axioms satisfied by DCScore, highlighting its role as a principled diversity evaluation method. Experimental results on synthetic datasets reveal that DCScore enjoys a stronger correlation with multiple diversity pseudo-truths of evaluated datasets, underscoring its effectiveness. Moreover, both empirical and theoretical evidence demonstrate that DCScore substantially reduces computational costs compared to existing approaches. Code is available at: https://github.com/BlueWhaleLab/DCScore.

📄 PDF Abstract BibTeX arXiv:2502.08512

Code (1)

bluewhalelab/dcscore 공식 구현 pytorch

Tasks

ClassificationDiversitytext-classificationText Classification

Similar Papers 제목 키워드 기반

Fidelity-Diversity Metrics for Text

2026-07-06 · Amanda Wang, Tudor Manole, Florentina Bunea, John Thickstun arxiv

As language modeling technology matures, there is an increasing research focus on the composition and curation of datasets used to train these models. For instance, practitioners commonly seek to augment high-quality dat…

Data Augmentation

Measuring diversity of synthetic prompts and data generated with fine-grained persona prompting

2025-05-23 · Gauri Kambhatla, Chantal Shaib, Venkata Govindarajan

Fine-grained personas have recently been used for generating 'diverse' synthetic data for pre-training and supervised fine-tuning of Large Language Models (LLMs). In this work, we measure the diversity of persona-driven …

Diversity

SynAE: A Framework for Measuring the Quality of Synthetic Data for Tool-Calling Agent Evaluations

2026-05-21 · Shuaiqi Wang, Aadyaa Maddi, Zinan Lin, Giulia Fanti arxiv

Today, tool-calling agents are commonly evaluated or tested on static datasets of execution traces, including input commands, agent responses, and associated tool calls. However, internal production datasets are often in…

A Penalty Goes a Long Way: Measuring Lexical Diversity in Synthetic Texts Under Prompt-Influenced Length Variations

2025-07-20 · Vijeta Deshpande, Ishita Dasgupta, Uttaran Bhattacharya, Somdeb Sarkhel 외 arxiv

Synthetic text generated by Large Language Models (LLMs) is increasingly used for further training and improvement of LLMs. Diversity is crucial for the effectiveness of synthetic data, and researchers rely on prompt eng…

Prompt Engineering

emb-diversity: A Tool for Embedding-Based Measurement of Data Diversity

2026-07-22 · Cantao Su, Menan Velayuthan, Esther Ploeger, Dong Nguyen 외 arxiv

There is growing evidence that data diversity is crucial for developing fair and robust NLP models. However, current approaches to measure diversity remain inconsistent and fragmented: While there exist a number of tools…