Datasheets Aren’t Enough: DataRubrics for Automated Quality Metrics and Accountability

ACL ARR 2026 January Submission106 Authors

Keywords: position paper, datasheets, datasets

Abstract: High-quality datasets are fundamental to training and evaluating machine learning models, yet their creation, particularly with accurate human annotations, remains a significant challenge. Many dataset submissions lack originality, diversity, or rigorous quality control, and these shortcomings are often overlooked during peer review. Submissions also frequently omit essential details about dataset construction and properties. Although existing tools such as datasheets promote transparency, they are largely descriptive and do not provide standardized or measurable methods for evaluating data quality. Likewise, metadata requirements at conferences encourage accountability but are inconsistently enforced. To address these limitations, this position paper advocates integrating systematic, rubric-based evaluation metrics into the dataset review process, especially as submission volumes continue to grow. We further explore scalable and cost-effective approaches to synthetic data generation, including dedicated tools and LLM-as-a-judge methods, to support more efficient evaluation. As a call to action, we introduce DataRubrics, a structured framework for assessing the quality of both human- and model-generated datasets. Leveraging recent advances in LLM-based evaluation, DataRubrics offers a reproducible, scalable, and actionable solution for dataset quality assessment, enabling both authors and reviewers to uphold higher standards in data-centric research.

Paper Type: Long

Research Area: Resources and Evaluation

Research Area Keywords: evaluation methodologies, evaluation

Contribution Types: Position papers

Languages Studied: English

Submission Number: 106